# AI New Zealand New Zealand's independent, vendor-neutral AI for business advisory, founded in 2017. Practical AI advice, training and tools for Kiwi organisations, including the AI Academy, the Great New Zealand AI Roadshow, executive briefings and the AI Operating System. --- # AI Advice, Solutions & Training for New Zealand Businesses Source: https://www.newzealand.ai/ Est. 2017 · Aotearoa # Everyone has *AI.* The difference is *you.* Some organisations are leaping ahead with AI enhancing their people & processes. Others are stuck, wondering where to start. **We help with practical advice, training, tools and custom solutions** built for people ready to work smarter, move faster and grow. [Book a Discovery Call](https://www.newzealand.ai/contact-us/) 3,542 AI Academy Members The gap ## Your team already has AI. Results are another story. Copilot is switched on. ChatGPT and Claude are ready to answer. A few people are doing genuinely impressive things, most aren't, and the gap widens every week. ### Scattered experiments. Everyone prompts their own way, so quality depends on who happens to be at the keyboard. Your team's best thinking stays trapped in individual chat histories. ### Pilots without owners. The demo impresses, then drifts. No named owner, no baseline, no target, so nothing moves on the P&L. That's AI theatre & potential left on the table. ### Risk nobody governs. Company data in personal accounts. No guardrails on what AI drafts, checks or decides. The exposure grows quietly while the policy sits in a drawer. Those who implement AI in a measured and managed way will outperform those who don't. The difference is in your people, and the questions you ask along the way. The NZ AI Benchmark ## What stage is your business at? *What's your next step?* Forty questions about a normal Tuesday in your organisation. Twenty minutes. Your stage on the AI Playbook's five, the trap at that stage, and the one move that takes you one stage up. Free, and no account needed to start. - **Against best practice, now.** Every dimension shows the stage you are at and what good looks like one stage up. - **Against other New Zealand organisations, as the cohort grows.** Peer comparison opens once 20 organisations in your industry, size band or region have taken it. - **Honest by design.** Not sure is a valid answer. It counts as a visibility gap, never as a low score. **40** questions **8** dimensions **5** stages **20** minutes [How the benchmark works →](https://www.newzealand.ai/benchmark/) ### Four quick answers to start. These make the benchmark possible. No name or contact details are collected here. Takes you straight to question one on the client platform. How we work ## A clear pathway, not another pilot. Start where it suits you. Each step stands alone, and each one sets up the next. 1 ### Executive AI Briefing. 3 hours · leadership team, max 15 Three hours with your leadership team to align on where AI creates the most value in your business. You leave with a high-level roadmap, our recommendations, and guardrails for safe, responsible use, so you move a metric, not a use case. [Learn more →](https://www.newzealand.ai/executive-ai-briefing/) 2 ### AI Lunch & Learn. 1 hour · your whole team A fast, practical session on what AI is, what it isn't, and how Kiwi businesses are using it right now. Live tool demos, real use cases and open Q&A. The quickest way to get everyone past the fear, over the hype & on the same page. [Learn more →](https://www.newzealand.ai/lunch-n-learn/) 3 ### Use Cases and Roadmap. moving metrics Your whole team submits use cases through your AI New Zealand client portal, where we review and prioritise them into a detailed, comprehensive Roadmap. Our readiness assessment and recommendations show your leadership why, where and how to invest in AI for measurable business impact. [Talk to us →](https://www.newzealand.ai/contact-us/) 4 ### AI Operating System. Built in weeks · improves every month Your methodology, templates and past work encoded once and shared by everyone, connected to Microsoft 365 and the tools your team already uses. Governed properly: AI drafts, humans approve. This is the stage where the value compounds. [Get started →](https://www.newzealand.ai/ai-os/) 5 ### AI Academy training. Ongoing · every role Weekly AI Toolkit webinars, the Microsoft Copilot, Claude at Work and AI Leaders certificate courses, and a growing library of resources and templates. Practical skills for the people who do the work, not just the keen early adopters. Capability across the whole team is what separates results from theatre. [Explore the Academy →](https://academy.newzealand.ai/) What is an AI OS? ## Built for how your business actually works. Five integrated layers that turn AI from a personal productivity tool into a company-wide operating capability. The order matters. Governance sits at the top and Skills at the bottom, because the thing most organisations buy first is the thing that should be built last. Layer 1 ### Governance layer. The policies and guidelines that set the guardrails for judgement and decisions. What your people are trusted to decide with AI, what has to come back to a human, and where the lines sit. AI drafts. Humans approve. Layer 2 ### Processes layer. Your standard operating procedures, owned and documented. Baseline metrics captured before anything changes, so every updated workflow shows measurable impact rather than a feeling that things are faster. Layer 3 ### Knowledge layer. Your methodology, brand voice, templates and past work encoded and reusable. Consistent quality across every team member, every time. Layer 4 ### Integrations layer. Connections to the sources of information your people need to complete a task. Microsoft 365, CRM, Teams, SharePoint and the tools your team already use every day. Layer 5 ### Skills layer. AI workflows configured for your most common output types: proposals, reports, meeting minutes, client communications, advisory documents. The shift Everyone is buying the skills layer. The future thinkers are investing in the *[integrations layer](https://www.newzealand.ai/integration-layer/)*. The AI OS is operational infrastructure. Designed to integrate data, process and people in one platform that can scale and flex with your business. Your firm's thinking, standards and past work become systemised and reusable, so the whole team produces high-quality work faster and more consistently. [Learn more about the AI OS](https://www.newzealand.ai/ai-os/) Proof ## You're in great company. Some of the organisations we're proud to work with, from national institutions to fast-moving mid-sized firms. [- Hunter Civil](https://www.huntercivil.co.nz) [- Your Way Kia Roha](https://www.yourwaykiaroha.nz) [- Aspire Property Management](https://aspireproperty.co.nz) - GWE - Gillies Group - Business Mentors - Research Review - AoFrio - WellingtonNZ - Just Cabins - Hospice - LawPlus - APD - Antipodes - PMG - NZFC - TEC - Motive - House Of Travel - Bayleys - Department of Conservation - Fire and Emergency New Zealand - Samsung - Rabobank Client case story ### [A virtual property manager, built into Aspire's own portal.](https://www.newzealand.ai/aspire-property-management-case-story/) An AI assistant inside Aspire's own management portal, grounded in its SOPs, the Residential Tenancies Act and six insurance policies. It drafts across email, WhatsApp and maintenance, and a property manager approves every message. Read the case story ~3,749 Hours a year released across four workflows $299,920 Indicative annual value of that time 180 Target properties per property manager, up from 120, no new hires 8,000+ Emails auto-drafted since August 2026 Certify by AI New Zealand · Self-paced certificate courses ## Four courses. Start with the foundations, then lead or learn the tools. AI at Work Foundations for everyone before they are handed the tools, AI Leaders for the person responsible for all of it, and Claude or Copilot, whichever your organisation actually runs on. All four are self-paced, built on New Zealand workplace scenarios, and end in a certificate from AI New Zealand. AI at Work Foundations ### What you should know *before working with AI.* The rules you work inside, what these tools are and where they fail, what they can already reach, and what stays your job. Written for New Zealand workplaces, and tool-neutral. **11** sections **5** practical exercises **Human-marked** assessment $235 pp + GST Lower rates for teams on a Business account [Start the course](https://certify.newzealand.ai/courses/ai-at-work-foundations) [See the full course →](https://www.newzealand.ai/ai-foundations-course/) AI Leaders ### Make AI safe to say yes to. *The AI Leaders course.* A self-paced certificate course for the person made responsible for AI across an organisation. Set the rules, bring your people with you, and finish with a ninety day plan you can defend in front of a board. **7** sections **7** practical exercises **Human-marked** assessment $495 pp + GST $295 pp + GST for 5 or more [Start the course](https://certify.newzealand.ai/courses/ai-leaders) [See the full course →](https://www.newzealand.ai/ai-leaders-course/) Claude at Work ### From Prompts to Projects. *The Claude at Work course.* Build an AI that knows your organisation, writes in your voice, follows your standards, and takes whole jobs off your desk while you keep the judgement. **4** sections **17** lessons About **3 hours** $235 pp + GST Lower rates for teams on a Business account [Start the course](https://certify.newzealand.ai/courses/claude-at-work) [See the full course →](https://www.newzealand.ai/claude-at-work/) Microsoft Copilot ### Microsoft Copilot *Certificate Course.* Become genuinely useful with Copilot: work safely inside your organisation's data, hand real jobs to Copilot in Word, Outlook, Forms, Excel, PowerPoint and Teams, and use Researcher, Analyst and custom agents. **4** sections **14** quizzes About **5 hours** $235 pp + GST Lower rates for teams on a Business account [Start the course](https://certify.newzealand.ai/courses/microsoft-copilot) [See the full course →](https://www.newzealand.ai/microsoft-copilot-course/) Training a whole team? Business, Corporate and Industry association accounts get their own course page and lower rates from the first seat. [Talk to us](https://www.newzealand.ai/contact-us/) or [set one up on Certify](https://certify.newzealand.ai/courses). Certificates issued by AI New Zealand. AI at Work Foundations and AI Leaders are awarded on a human-marked final assessment. Not affiliated with, or endorsed by, Microsoft, Anthropic or any other AI vendor. AI Academy ## Upcoming webinars. Live, practical sessions every week in the AI Academy. Learn the tools, see them in action and ask your questions. AI Toolkit webinars are for [Premium Members](https://www.newzealand.ai/premium/). FRI 25 Sep · 11:00am NZT · 45 min ### Build Workflow Two: The Project Activity Tracker AI Toolkit webinar · Justin Flitter 90 registered [Register →](https://academy.newzealand.ai/c/the-ai-toolkit-webinars/workflow-series-2) FRI 9 Oct · 11:00am NZT · 45 min ### Build Workflow Three, Together AI Toolkit webinar · Justin Flitter 90 registered [Register →](https://academy.newzealand.ai/c/the-ai-toolkit-webinars/workflow-series-3) FRI 16 Oct · 11:00am NZT · 45 min ### Deploy All Three, For Real AI Toolkit webinar · Justin Flitter 90 registered [Register →](https://academy.newzealand.ai/c/the-ai-toolkit-webinars/workflow-series-4) [Join the Academy](https://academy.newzealand.ai/) [See all events →](https://academy.newzealand.ai/c/events/) Insights ## Our latest articles. *More in the Academy.* Research, insights and best practice for companies making the move. Client case story ### A virtual property manager, built into Aspire's own portal. [Read article →](https://www.newzealand.ai/aspire-property-management-case-story/) AI Operating System ### Everyone is buying the skills layer. Almost nobody is buying the integration layer. [Read article →](https://www.newzealand.ai/integration-layer/) 2026 AI Roadshow Notes ### AI is no longer just about content generation. It is becoming operational infrastructure. [Read article →](https://www.newzealand.ai/operational-infrastructure/) [Read all articles →](https://www.newzealand.ai/insights/) [Read more in the Academy →](https://academy.newzealand.ai/c/insights) Who's behind AI New Zealand ## Champions of AI for business since 2017. Justin Flitter founded AI New Zealand in 2017 to help business leaders discover why, where and how to leverage Artificial Intelligence. Since then the team has helped thousands of Kiwi organisations with practical advice and training, from leadership tables to workshop floors across the country. "Adoption automates today's work. Value lives in redesigning it." [Book Justin to speak](https://keynote.newzealand.ai) [Book an AI Chat](https://www.newzealand.ai/contact-us/?service=ai-chat#book) Our impact ## 9 years and counting. We started ahead of the curve. Now we help organisations across Aotearoa leverage this rapidly changing technology to work smarter, move faster and grow. Advice at the leadership table. Training for the people doing the work. And implementation that embeds AI into everyday workflows. We're proud to work with teams of five to workforces of 500 plus in forward-thinking organisations across Aotearoa. Including professional services, construction, manufacturing, engineering, not-for-profit and government organisations. Client verified metrics 484 Client engagements 495 Use cases evaluated 48,383 Annual hours saved $6.84M Annual costs saved Updated September 2026 · verified from the AI New Zealand client platform. Hours and cost saved are counted from live use cases only. --- # The AI Enablement Playbook | AI New Zealand Source: https://www.newzealand.ai/ai-playbook/ The AI enablement playbook # The playbook for AI your whole business *actually uses.* Six steps, in order, from a first safe win to work redesigned around what AI makes possible. This is the route we take clients through, the free toolkits that do the heavy lifting, and the platform that keeps the whole thing measured. See the six steps Where should you start? Where adoption starts ## Six places Kiwi businesses get their first wins. Start where the work is repetitive, the value is obvious and the risk is low. These six earn belief across the team and buy you the permission to go deeper. ### Meeting management. Transcribe, summarise and capture actions in real time, so decisions are documented and follow-ups are clear. ### Risk & compliance. Monitor policies, compare documents and flag anomalies in compliance-heavy processes, consistently and auditably. ### Content production. Draft, repurpose and stay on brand across blogs, social media and internal comms. ### Access to knowledge. Search across emails, documents and systems for fast, context-aware answers. A company-wide knowledge concierge. ### Proposals & presentations. Generate outlines and tailor content to each client, so your people focus on insight and impact. ### Research & decision making. Analyse trends, summarise research and simulate scenarios, turning complexity into clarity for leaders. Get one of these working properly before you attempt six. The shift ## Adoption is the on-ramp. Adaptation is the destination. Most AI strategies are adoption strategies wearing a transformation badge. Adoption is using AI to do today's work faster. You bolt a tool onto a workflow and shave time off a task. The workflow itself does not change, and that workflow was designed around legacy tech and legacy assumptions you no longer have to honour. Adaptation is redesigning the work itself. It asks a braver question: if AI were assumed from day one, how would we design this at all? Here is the trap. Most of the value lives in adaptation. Most of the effort goes into adoption. So businesses end up with better screenshots and the same bottlenecks. That is AI theatre, and it is the most expensive place to get comfortable. "If your AI rollout is making your old workflow faster, you are still doing adoption. No matter how many tools you have deployed." ### Adoption. Speeds up the work you already do. Task-level. Measured by time saved and tools deployed. Necessary, but it does not change your competitive position. ### Adaptation. Redesigns how the work is done. Operating-model level. Measured by cycle time, cost-to-serve and new capability. This is where value compounds. You need both. But only adaptation changes the game. First principles ## Three principles that hold the whole thing together. ### Critical thinking is your competitive moat. As AI commoditises answers, the real advantage is human judgement, context, and the courage to challenge both the model and the status quo. Critical thinking turns generic AI outputs into distinctive decisions your competitors cannot easily copy. ### Move a metric, not a use case. Anchor every initiative to a value lever: revenue, cost-to-serve, cycle time, risk or customer experience. When AI is tied to a metric that matters, you stop doing AI theatre and start delivering measurable business impact. ### Redesign your operating logic. AI should not sit on the edge of your business. It should reshape how you sense, think, decide and act every day. Make that operating logic explicit and AI becomes trusted infrastructure, not another shiny experiment in the innovation corner. The playbook ## Six steps. Two altitudes. Done in order. These are the six moves that take a business from scattered experiments to AI used properly, and the same six, done deeper, that carry you into adaptation. The mistake is treating them as a checklist you finish once. They are a discipline you keep levelling up, and each one has something practical attached to it. 1 ### Assign an AI Leader. Start here One person leads education and adoption, gets tools into hands safely, and gives nervous staff their first win. Make time for the person who knows your customers and your process, and bring them with you. Then go deeper The AI Leader becomes an AI council and a centre of enablement. Leadership drives it together, aligned to the business plan. This is not delegated: federating it out to run independently does not scale, the parts clash later. Use this Our [AI Leaders Course](https://certify.newzealand.ai/courses/ai-leaders) gives the person you choose a mandate, a register, a costed plan and a risk position they can defend. 2 ### Set governance protocols. Start here A clear, organisation-wide AI policy. A statement of what you will and will not do with AI, so staff can make confident, responsible decisions. Then go deeper Governance grows into agent assurance. Once agents act in production, and talk to other agents, you need a quality system: sampling, audit, and assurance profiles that tighten when an agent drifts. If you are talking about AI and not talking about cyber, you are in trouble. Use this The policy sections and the tool approval framework are in the [Champions pack](https://www.newzealand.ai/ai-champions-committee/). Clients build and keep theirs in the [Client Engagement Platform](https://www.newzealand.ai/platform/). 3 ### Identify practical use cases. Start here Workshops to find quick wins that boost productivity. Focus on a few repetitive, time-consuming tasks to build capability, get value on the board fast and build belief. Then go deeper Use cases become workflow redesign. You map how a critical workflow senses, thinks, decides and acts, then rebuild it around AI rather than bolting AI on. You move a metric, not a use case. Use this Your whole team logs opportunities in the [Client Engagement Platform](https://www.newzealand.ai/platform/), scored and ranked into a delivery order with the hours and dollars attached. 4 ### Understand your risks. Start here Stand up an AI committee. Review your data, IP, security and privacy against how employees could use AI tools, so AI use stays safe and compliant with regulatory requirements and industry standards. Then go deeper Risk becomes a leadership discipline: setting an explicit risk appetite for how close to the frontier you will operate, and owning assurance so problems are caught before they end up in the papers. Prompts are not controls. Real controls are technical. Use this Three hours with your leadership team in an [Executive AI Briefing](https://www.newzealand.ai/executive-ai-briefing/) is where risk appetite gets set, written down and owned. 5 ### Select the right AI tools. Start here A selection framework that prioritises security, scalability and integration with your data, and favours tools that serve multiple use cases. Choose well, avoid the shiny-object trap. Then go deeper Tools become a system: a shared [AI operating system](https://www.newzealand.ai/ai-os/) with a governance layer, a processes layer, a knowledge layer, an integrations layer and a skills layer. And a system is run, not installed, so it is maintained and improved every month. Use this The free [Organisational Brain Toolkit](https://www.newzealand.ai/organisational-memory/) builds the knowledge layer first: eight file templates, about an hour to a working set. No tool is much use without it. 6 ### Provide ongoing AI skills training. Start here Practical, hands-on training so the team can use the tools and operationalise new workflows. Build confidence and capability, and power up fast adopters with more advanced skills. Then go deeper You build an AI-native workforce. Roles deepen, they do not disappear. Frontline staff own agent quality, experts move from firefighting to calibrating, and leaders invest in literacy across the whole organisation. Use this The [AI Academy](https://academy.newzealand.ai/) is free to join: weekly AI Toolkit webinars, the Microsoft Copilot course and a growing library of templates. Same six steps, twice. The second pass is where the value is. Our toolkits ## Four of the six steps come ready-made. Two toolkits, covering steps 1, 2, 4 and 5 between them. We built them for clients and then gave them away, because the businesses that get AI right are the ones who do this work early. Both are free, and both are editable files rather than a PDF you can only read. Steps 1, 2 and 4 · lead it, govern it, know the risks ### The AI Champions Committee Resource Pack. The complete programme for standing up AI governance and adoption inside a 50 to 250 person business. You need a sponsor with authority, seven to nine people who will turn up, and ninety days. - A 115 page pack, plus every template as an editable file - Terms of reference, seat map and the champions network - Use case register, tool approval framework and policy sections - A ninety day implementation plan, with an output at each stage [Get the pack →](https://www.newzealand.ai/ai-champions-committee/) Step 5 · the knowledge layer ### The Organisational Brain Toolkit. Your team briefs AI from zero every single time. The fix is not a data project and it is not a new tool: it is about eight plain documents, written once, pointed at the AI your team already has. - Eight file templates, from company profile to definition of done - About an hour to a working set, dictated rather than written - The 65 minute method for getting a process out of an expert's head - Setup checklists for Claude Projects and Copilot notebooks [Get the toolkit →](https://www.newzealand.ai/organisational-memory/) There is a third, on getting AI to apply your fonts, colours and layout to everything it produces. All of them live on the [toolkits page](https://www.newzealand.ai/toolkits/). The system underneath ## Adaptation needs a system, not more tools. The second pass through the playbook does not arrive because everyone got better at prompting. It arrives when the way your business works is encoded once, connected to your real systems, used by everyone, and improved every month. That is an AI operating system, and it is our flagship managed service. ### Governance layer. The policies and guidelines that set the guardrails for judgement and decisions: what your people are trusted to decide with AI, what comes back to a human, and where the lines sit. With clear measures for time recovered and adoption. AI drafts. Humans approve. ### Processes layer. Your standard operating procedures, owned and documented, with baseline metrics captured before anything changes, so every updated workflow shows measurable impact rather than a feeling that things are faster. ### Knowledge layer. Your methodology, templates, brand voice and past work encoded and reusable, so quality does not depend on who happens to be at the keyboard. ### Integrations layer. Connected to the systems you already run: Microsoft 365, your CRM, Teams and SharePoint. Less manual filing, less context switching, more work that finishes itself. ### Skills layer. The outputs your team produces every week, proposals, reports, client communications, minutes and advisory documents, configured once as AI workflows and available to everyone. Most teams are live in five to nine weeks, from discovery to first workflows. [Explore the AI OS](https://www.newzealand.ai/ai-os/) [Book a discovery call](https://www.newzealand.ai/contact-us/) How we keep it measured ## Every engagement comes with a workspace. A playbook nobody can see the progress of is a document. The Client Engagement Platform is where your version of this lives: your objective, your use cases, your delivery plan, your governance documents and the value the work is returning. We designed and built it ourselves, and it is live from the day we start. - Hours won back and dollars saved across your use cases, updating as each one goes live. - A register your whole team fills in, scored and ranked into a delivery order. - Delivery phases, deliverables and dates, so you can see what is under way and what is next. - Your AI policy, guidelines and documents, kept in one secure place. - Access you set yourself, with a one time email code instead of a password. [Take the tour](https://www.newzealand.ai/platform/) [Book a discovery call](https://www.newzealand.ai/contact-us/) Where do you start? ## Pick the one that sounds like a normal Tuesday. Be honest, not aspirational. The right first move depends far less on your industry than on which of these five sentences is true in your business this week. **Start at step 1, with shared understanding.** This is the easiest move on the whole playbook and the gap to your competitors widens every week you skip it. An [AI Lunch & Learn](https://www.newzealand.ai/lunch-n-learn/) gets your whole team past the fear and the folklore in an hour, and gives your AI Leader something to build on. **Go straight to steps 2 and 4, governance and risk.** This is the riskiest place to sit: ungoverned tools, unshared wins, and your best people quietly feeding company context into something nobody is watching. An [Executive AI Briefing](https://www.newzealand.ai/executive-ai-briefing/) sets the policy, picks the value levers and gives the experiments a direction. The [Champions pack](https://www.newzealand.ai/ai-champions-committee/) is free and you can start this week. **You are at the adoption ceiling. Step 3, done properly.** You have made the legacy way of working faster without asking whether the legacy way is still the right way. The move is to pick one core workflow, redesign it around AI, and put a measurable target on it. That is where the [AI OS](https://www.newzealand.ai/ai-os/) conversation usually starts. **Second pass, all six steps.** You are past the point most businesses never reach. The work now is depth and rhythm: more workflows redesigned, capability lifted every month, governance that scales with you, and someone accountable for the numbers. That is exactly what the [AI Operating System](https://www.newzealand.ai/ai-os/) and the [Client Engagement Platform](https://www.newzealand.ai/platform/) are for. **Rare air. Step 2, at the top of its range.** The discipline now is assurance: agent quality, an explicit risk appetite, and an operating rhythm to stay at the edge responsibly. Run as a managed service, that rhythm is monthly, not annual. Let us compare notes. Wherever you landed, the next move is the same shape: one step, properly, not six at once. [Book an Executive AI Briefing](https://www.newzealand.ai/executive-ai-briefing/) How we help ## Wherever you start, there is a next step. Get everyone started ### AI Lunch & Learn. Demystify AI, build confidence, and get tools into hands safely. The fastest way to give your whole team a shared starting point. [Learn more →](https://www.newzealand.ai/lunch-n-learn/) Set the direction ### Executive AI Briefing. Three hours with your leadership team to uncover high-impact use cases, set governance, and leave with an AI Roadmap. [Learn more →](https://www.newzealand.ai/executive-ai-briefing/) Find the value ### Use Case Scoping. Find the highest-value opportunities, scope them properly, and quantify the ROI before you invest. The business case for AI, on one page. [Talk to us →](https://www.newzealand.ai/contact-us/) Redesign the work ### AI Strategy. A company-wide, six to nine week programme that anchors AI to your business goals and shows you how to redesign, not just speed up. [Talk to us →](https://www.newzealand.ai/contact-us/) Flagship Run it as a system ### AI Operating System. Our managed service, and the engine of adaptation. We build your governance, processes, knowledge, integrations and skills layers, get your first workflows live in five to nine weeks, then run and improve the system with you every month. [Explore the AI OS →](https://www.newzealand.ai/ai-os/) Everything, in one document ### Request our Client Services Pack. The full menu: every service, format and price in one document. Email us and we will send it straight over. [Request the pack →](mailto:hello@newzealand.ai?subject=Please%20send%20me%20your%20Client%20Services%20Pack) You do not have to do this alone, and you do not have to do it all at once. The next move ## The tools are ready. *Are You?* Those who implement AI in a measured and managed way will outperform those who do not. Let us help you find the one workflow worth redesigning first, then build the system that keeps you improving. [Book an Executive AI Briefing](https://www.newzealand.ai/executive-ai-briefing/) [Explore the AI OS](https://www.newzealand.ai/ai-os/) [Request our services](https://www.newzealand.ai/contact-us/#request-services) --- # The AI Operating System that gets work done | AI New Zealand Source: https://www.newzealand.ai/ai-os/ AI-powered business operating system # Redesign how you create value. *With AI in the loop.* Stop relying on individuals using AI differently. Give your team a shared, managed AI operating system: built for your business, connected to your tools, and maintained to keep improving. [Book a discovery call](https://www.newzealand.ai/contact-us/) See pricing ## Why teams switch. $15k+ indicative value recovered per person, per year 5–9 weeks from discovery to first workflows live in your team Managed. Measured. Consistent. one shared system, owned and improved every month Repetitive tasks, delegated. Jobs done. The evolution of AI in business ## Most organisations are stuck between experimentation and scale. Individual tools. Inconsistent outputs. Knowledge trapped in whoever happens to be using AI that day. That's AI theatre: it looks productive, and it doesn't move a metric. The next step isn't more tools. It's an operating model. One that encodes how your firm works, connects AI to your systems, and gives every person on your team a trusted AI assistant they can delegate work to, consistently. 1 ### Individual experimentation. People discover AI individually. Results vary wildly. No shared standards. 2 ### Policies and basic training. An AI policy lands. A workshop runs. Some teams get curious, others stay sceptical. 3 ### Scattered tool adoption. Copilot, Claude, ChatGPT, Perplexity. Everyone using different tools. Highest risk: hard to govern, harder to measure. The move 4 ### A shared AI operating system. Your methodology, brand voice and workflows encoded into a single shared AI environment. Connected. Governed. Improving monthly. The OS stage is where value compounds. What is an AI OS? ## Built for how your business actually works. Five integrated layers that turn AI from a personal productivity tool into a company-wide operating capability. The order matters. Governance sits at the top and Skills at the bottom, because the thing most organisations buy first is the thing that should be built last. Layer 1 ### Governance layer. The policies and guidelines that set the guardrails for judgement and decisions. What your people are trusted to decide with AI, what has to come back to a human, and where the lines sit. Permissions, versioning and the audit trail are how that policy is enforced. AI drafts. Humans approve. Layer 2 ### Processes layer. Your standard operating procedures, owned and documented. Baseline metrics captured before anything changes, so every updated workflow shows measurable impact rather than a feeling that things are faster. Layer 3 ### Knowledge layer. Your methodology, brand voice, templates and past work encoded and reusable. Consistent quality across every team member, every time. Layer 4 ### Integrations layer. Connections to the sources of information your people need to complete a task. Files, systems and applications: Microsoft 365, CRM, Teams, SharePoint and the tools your team already use every day. Layer 5 ### Skills layer. AI workflows configured for your most common output types: proposals, reports, meeting minutes, client communications, advisory documents. The shift Everyone is buying the skills layer. The future thinkers are investing in the *integrations layer*. The AI OS is operational infrastructure. Designed to integrate data, process and people in one platform that can scale and flex with your business. Your firm's thinking, standards and past work become systemised and reusable, so the whole team produces high-quality work faster and more consistently. [Book a discovery call](https://www.newzealand.ai/contact-us/) Who it's for ## Built for knowledge-work organisations. Professional services, advisory, accounting, legal, financial services, property, and not-for-profits with a heavy documentation load. Best fit ### 10–20 person professional services firms. Big enough to feel the pain of inconsistency. Small enough to move fast. Consulting, advisory, accounting, legal, financial services or property. Strong fit ### 20–50 person firms ready for the next step. You've run the workshops. Copilot is live. Now you want a structured, ongoing operating model, not another round of prompting tips. This is the next step. Strong fit ### Local GOVT, Non-profits and industry associations. High documentation load. Compliance requirements. Knowledge held in long-serving staff. Your AI OS retains and systemises that institutional knowledge. AI OS managed service ## From discovery to daily use in five to nine weeks. Six steps. We start inside your existing policies, and finish with your team using the system every day. The five to nine week range is set by you, not us: how quickly we get time with your people, and how quickly decisions get made at the checkpoints. Then we manage, improve and expand it every month on retainer. The system gets more valuable over time. 1 ### Discovery. Purpose, people and process. We define how AI enablement anchors to your strategy and objectives, and review your AI policy, privacy, security and data handling rules so we know the contractual, data and technology guardrails the programme has to work inside. We map where AI could add the most value, and document your data sources, systems, access rights and integration options. **You get an AI Opportunities report**, and a checkpoint where you decide whether to focus or expand before the next stage. 2 ### Process mapping. Tasks, workflows and procedures. Your SOPs, process workflows or recorded interviews documented, with baseline metrics captured for benchmark reporting. Each process is recorded in the Use Case Register and assessed against business impact, technical complexity and time to value. **You get a prioritised roadmap of use cases**, and you select around five to build first. 3 ### Architecture and skill design. Knowledge and integrations. Which sources of information each task needs, and which outputs are worth building for. We design the connector architecture, the skill set and the private plugin environment, with every decision tied back to a workflow and a KPI. **You get an AI Sprint for each use case**, scoping the detail with your process owners. 4 ### Build, deploy and test. Connections deployed and Skills built against the agreed design. Every Skill is validated on your own documents and data, and against your guardrails, before your team touches it. **You get a working environment.** 5 ### Training, rollout and handover. Team training sessions, usage guidance and launch support. Your people know the rules as well as the tools. Administrator handover, so the system has an owner inside your business from day one. **You get a team that is confident and using it.** 6 ### Continual development. System administration, new staff training, one use case developed and deployed each month, and feedback loops for skill optimisation. **Every month you get** new Connectors and Skills, a 60-minute AI Sprint, a training session and a report. Included, not extra ## Your AI OS comes with a client workspace. The Client Engagement Platform is where your engagement lives: your objective, the use cases your team logs, the delivery plan, the governance documents and the value the work is returning. We designed and built it ourselves, and it is live from the day we start. - Hours won back and dollars saved across your use cases, updating as each one goes live. - A register your whole team fills in, scored and ranked into a delivery order. - Delivery phases, deliverables and dates, so you can see what is under way and what is next. - Your AI policy, guidelines and documents, kept in one secure place. - Access you set yourself, with a one time email code instead of a password. [Take the tour](https://www.newzealand.ai/platform/) [Book a discovery call](https://www.newzealand.ai/contact-us/) Pricing ## Simple, transparent pricing. A base engagement to build the system in Numa, Claude Cowork, Microsoft Copilot, or open-source models hosted on your own infrastructure, then a retainer to run and improve it. Rates are estimates only and exclude GST. ### Numa by Arcanum. From $20,000 Customised deployment of Numa, plus a retainer from $3,500 per month (minimum 3 months). - Suitable for teams of 5 to 100+ - Discovery and workflow audit - Use Case Roadmap - All five layers: governance, processes, knowledge, integrations and skills - Around 5 use cases, selected from a prioritised roadmap, with the data connections they need - Team training session - Documentation and operating guidance - Administrator handover - Every month on the retainer - New Connectors and Skills added monthly - 1 x 60-minute AI Sprint per month - Monthly training sessions - Monthly reporting [Get started](https://www.newzealand.ai/contact-us/?service=discovery-call&topic=ai-os#book) 4 Activations In Progress ### Claude Cowork. From $27,500 Managed service deployment of Claude Cowork, plus a retainer from $3,500 per month (minimum 6 months). - Suitable for teams of 20 to 100+ - Stakeholder discovery sessions - Workflow and document audit - Use Case Roadmap - All five layers: governance, processes, knowledge, integrations and skills - Around 5 use cases, selected from a prioritised roadmap, with the data connections they need - Private Claude plugin deployment - Team training and documentation - Administrator handover and launch support - Every month on the retainer - New Connectors and Skills added monthly - 1 x 60-minute AI Sprint per month - Monthly training sessions - Monthly reporting [Get started](https://www.newzealand.ai/contact-us/?service=discovery-call&topic=ai-os#book) ### Microsoft Copilot. From $32,500 Microsoft Copilot deployment. Plus a $5,000 per month managed service retainer (minimum 6 months). - Suitable for companies with 20 to 200+ users - Stakeholder discovery sessions - Workflow and document audit - Data & security audit - Use Case Roadmap - All five layers: governance, processes, knowledge, integrations and skills - Around 5 use cases, built as Agents or Skills from a prioritised roadmap, with the data connections they need - Includes Copilot Cowork - Includes Copilot Studio - Team training and documentation - Administrator handover and launch support - Every month on the retainer - New Connectors, Agents or Skills added monthly - 1 x 60-minute AI Sprint per month - Monthly training sessions - Monthly reporting [Talk to us](https://www.newzealand.ai/contact-us/?service=discovery-call&topic=ai-os#book) ### Open Source. From $32,500 Open-source models, such as Qwen, hosted on your own infrastructure. Plus $3,500 per month managed service (minimum 6 months). - Stakeholder discovery and workflow audit - Open-weight models hosted in your environment - Your data never leaves your infrastructure - All five layers: governance, processes, knowledge, integrations and skills - Around 5 use cases, selected from a prioritised roadmap - 2–3 system connectors - Team training and documentation - Administrator handover and launch support - Every month on the retainer - New Connectors and Skills added monthly - 1 x 60-minute AI Sprint per month - Monthly training sessions - Monthly reporting [Talk to us](https://www.newzealand.ai/contact-us/?service=discovery-call&topic=ai-os#book) Not sure which fits? Book a discovery call and we'll recommend the right starting point for your team. Tool selection ## The right platform for your team. We're platform-agnostic. The operating model comes first, then we deploy it on the platform that fits your team, your systems and your budget. NZ-built · fast to deploy ### Numa. A homegrown agent platform from Wellington's Arcanum. Pre-built AI agents take repetitive busywork off your team: quoting, reporting, meeting notes, support email and knowledge search. Fast to stand up, priced for SMEs, and already working inside Kiwi businesses. [asknuma.ai →](https://asknuma.ai/) Most popular ### Claude Cowork. Anthropic's Claude, set up as your shared AI environment: your methodology, brand voice and workflows encoded as skills, connected to your systems through a private plugin deployment. The platform behind most of our AI OS builds. Microsoft 365 organisations ### Microsoft Copilot. For organisations committed to the Microsoft ecosystem: agents and workflows built into Microsoft 365, Teams and SharePoint, with the governance controls your IT team already manages. Sovereign · runs on your infrastructure ### Open Source / Sovereign Infrastructure. Open-weight models such as the Qwen 14B model (or similar) hosted entirely on your own infrastructure. Your data never leaves your environment: ideal for organisations with strict data sovereignty, privacy or compliance requirements. We configure, deploy and manage it as a private AI environment. Good to know ## Frequently asked questions. ### What if the team won't adopt it? Nobody gets this sprung on them. Bringing your people along is part of the work: we build skills across the business, not just in one team. ### What if we're locked in? You're not. Our methodology is transferable, and it moves with you to platforms like ChatGPT, Microsoft Copilot, Numa and Claude. ### What happens if we cancel after six months? Nothing breaks. By then your people can run the AI OS themselves, and no part of it depends on us. We're here if you want a hand. ### What do we get with the monthly retainer? A full managed service: support, training and new AI-powered workflows every month. Your AI OS gets more capable, and more valuable, as it goes. See what each tier includes → Something we haven't covered? [Book a discovery call](https://www.newzealand.ai/contact-us/) and ask us directly. --- # The Client Engagement Platform | AI New Zealand Source: https://www.newzealand.ai/platform/ Included in every engagement # Your AI Enablement program. *Planned, delivered & measured.* Every organisation we work with gets the Client Engagement Platform from the day we start. Your objective, your use cases, your delivery plan, your governance documents and the value the work is returning, in one place your whole team can see. [Book a discovery call](https://www.newzealand.ai/contact-us/) Take the tour ↓ Take the tour ## One platform. Built with AI. Designed for the way we work. Not a mock up. This is the platform as your team would use it. Totara Bay Consulting is a demonstration account. **The value you are seeing so far**Hours won back and dollars saved across your use cases, updating as each one goes live. Estimated and measured figures are labelled as such, so nobody has to guess which is which. **A register your whole team fills**Anyone with a work email at your approved domains can log an AI opportunity. You score them, prioritise them, and rank the approved work into a delivery order. **Delivery you can follow**Your program phases, deliverables and dates. What is under way, what is finished, what is coming next, and what we are waiting on you for. **Governance that holds up**Your AI policy and guidelines, built from your answers and written to be enforced and defended, with the New Zealand legal framework named rather than gestured at. **Scoped on facts, not assumptions**The software your business actually runs on, and the AI platforms already in play, each marked critical, required or replaceable. That is what integration work gets scoped against. **You control the access**Give a colleague full access, or team access to just the areas you tick. Everyone signs in with a one time email code, so there are no passwords to manage. Why it exists ## AI programs lose momentum in the gaps. Not in the work. In the space between the pieces of it, where nothing is anyone's job. ### The roadmap is in one person's inbox. So the answer to what happens next depends on whether that person is at their desk. ### The policy is in a shared drive. So it gets written once, filed, and never read by the people it was written for. ### The numbers are in a spreadsheet. So when the board asks what the AI program returned, the honest answer takes a fortnight. We designed and built the platform ourselves, for the way we deliver Activation Programs. It keeps those things together, so your team works from the same picture and progress stays visible to the people who need to see it. Your workspace is live from the day we start. Security and control ## Your workspace, your rules. Access is scoped to your organisation, and you decide who in your team sees what. ### No passwords to manage Everyone signs in with a one time code sent to their work email. Nothing to reset, nothing to leak. ### Access you set yourself Full access for the people running the program, or team access limited to the areas you tick, person by person. ### Published, not promised Our platform terms, privacy policy, data processing agreement and full sub-processor list are [published in the open](https://clients.newzealand.ai/legal). [Book a discovery call](https://www.newzealand.ai/contact-us/) [Client sign in](https://clients.newzealand.ai) Included, not extra ## See it with your own engagement in it. The platform comes with every Activation Program at no extra cost. Book a discovery call and we will walk you through the workspace, and what your first ninety days in it would look like. [Book a discovery call](https://www.newzealand.ai/contact-us/) [Explore the AI Operating System](https://www.newzealand.ai/ai-os/) --- # Executive AI Briefing | AI New Zealand Source: https://www.newzealand.ai/executive-ai-briefing/ Executive AI Briefing · 3 hours · leadership team # Executive AI *Briefing.* Empower your leadership team to make informed decisions on why, where and how to leverage Artificial Intelligence. Discover how to leverage AI powered technology to boost your competitiveness, productivity and innovation. [Book an Executive AI Briefing](https://www.newzealand.ai/contact-us/) See what's included **Max 15** directors and senior leaders **$5,500** + GST and travel **Online** or in-person About this program ## About this program. We facilitate Executive AI Briefing sessions for senior leadership teams in businesses and organisations across Aotearoa. 1 Pre-workshop planning call 2 Pre-workshop survey for all participantsTheir responses shape the day 3 3 hour workshop delivered online or in-person 4 1–3 interviews with employees to identify practical use cases 5 Post workshop you'll receive a report outlining high-level use cases, potential solutions, organisational enablement recommendations and next steps for your AI Roadmap ### Terms & conditions. - 👉 Limited to 15 participants - 👉 Attendees must be your directors or senior leaders - 👉 Recordings are only for internal use and must never be published or shared externally - 👉 Cost: **$5,500 plus GST** and travel costs The briefing ## What you'll learn. This three hour briefing will inform, inspire and entertain, showcasing the capability of AI today with time for discussions. Understand what AI is, the types of AI, use cases, capabilities and limitations Practical use cases and real-world applications Establish the ROI for AI. Discuss the uses for AI that will deliver quick wins and long term business value The AI for business tools and platforms and how to choose the right tools for your business Good governance protocols to transparently communicate how the business is using AI and inform decisions on responsible use Risk identification and management to reduce IP leakage and manage data privacy Leadership mindset: tackle common challenges and empower teams for the AI era [Book an Executive AI Briefing](https://www.newzealand.ai/contact-us/) Let's start today ## Power up your potential with practical AI skills. Join our community of ambitious leaders and take your career to the next level. [Book an AI Chat](https://www.newzealand.ai/contact-us/?service=ai-chat&topic=exec-briefing#book) [Request our services](https://www.newzealand.ai/contact-us/#request-services) --- # The NZ AI Benchmark: how does your business stack up? | AI New Zealand Source: https://www.newzealand.ai/benchmark/ The NZ AI Benchmark · Free · 20 minutes # How does your business *stack up?* A free 20 minute benchmark for New Zealand organisations. Forty questions, eight dimensions, and your stage on the AI Playbook's five, with the one move that takes you one stage up. Not five. [Start the benchmark →](https://clients.newzealand.ai/benchmark) How the benchmark works **40** questions **8** dimensions **5** stages **Peer** comparison as the cohort grows Your stage and profile Stage 3 · Adopting Solid line is you. Dashed line is one stage up.Sample profile How it works ## Twenty minutes to know where you stand. No account needed to start. Tell us four things about your organisation, answer honestly, and pick the row that sounds like a normal Tuesday, not the row you aspire to. Not sure is a valid answer: it counts as a visibility gap, never as a low score. 1 ### Answer forty questions. Five per dimension, each a plain description of what a normal week looks like. Role, size, industry and region first, so your answers can be benchmarked. 2 ### See your stage. Your stage on the AI Playbook's five, the trap at that stage, a radar of all eight dimensions and what you cannot see yet. Straight away. 3 ### Get the full report and the next move. Leave your work email for the detailed breakdown: commentary for every dimension, the patterns in your answers, how you stack up against other New Zealand organisations, and the one move that takes you one stage up. The framework ## Eight dimensions, six steps. The AI Playbook's six steps, plus leadership and the knowledge that feeds every tool. The tool question is the last question, so it carries the least weight. [Read the five stages and the six steps in full](https://www.newzealand.ai/benchmark/framework/). Leadership ### Strategy & Leadership Is there a number AI is meant to move, and does a leader own it? Step 1 ### People & Culture Is there an AI Ambassador, and does one person's improvement become everyone's baseline? The context engine ### Data & Knowledge Is how you do your best work written down somewhere an AI can use it? Step 5 ### Tools & Systems Managed accounts, an approved list people can name, and tools that add up to a system. Step 3 ### Use Cases & Workflows Has one workflow been redesigned around AI, and where does the saved time go? Step 2 ### Governance & Protocols Who decides, how fast, and is the data boundary enforced by tools rather than trust? Step 6 ### Skills & Training A baseline before tool access, training on real work, and a verification habit. Step 4 ### Risk, Privacy & Responsible AI Existing New Zealand law, disclosure and consent, Māori data as practice, and agents with checkpoints. Five stages ## One letter is the whole game. Most New Zealand organisations sit between Experimenting and Adopting and believe they are further along. The gap that matters is between stage 3 and stage 4: making the old work faster, or redesigning the work itself. Adopting or adapting. Stage 1 ### Aware Leaders reading headlines. **The trap:** mistaking awareness for progress. Stage 2 ### Experimenting Ad hoc use, no policy, no standard. **The trap:** this is the highest risk stage, not the safest. Stage 3 ### Adopting Policy, training and tools live. Work is faster. **The trap:** the adoption ceiling. Stage 4 ### Adapting Workflows redesigned around AI. Operating logic has changed. **The trap:** adapting everything at once. Stage 5 ### Frontier Agents run adapted workflows in production. **The trap:** forgetting the discipline. 87 percent of New Zealand organisations use some form of AI and only 12 percent have scaled it across the whole business (Datacom State of AI Index 2025). The distance between those two figures is the Adaptation Gap, and it is where the benchmark places you. The benchmark ## Against best practice now. Against other New Zealand organisations as the cohort grows. ### Best practice, from day one. Every dimension shows the stage you are at, the stage above it, and what good looks like there: a named number with an owner and a baseline, low risk decisions inside five working days, a training baseline before tool access, agents with a folder rather than a drive. ### Your peers, once there are enough of them. When at least 20 organisations in your industry, size band or region have completed the benchmark, you see where you sit among them. Until then you see how many have taken it so far. Respondents to the benchmark, not a representative sample, and we say so on the page. Free, and the next move is always one stage up ## Find out how you stack up. Then make one move. Take the benchmark now, or book a free CEO briefing call and we will walk through your stage, the pattern in your answers, and the one or two moves worth making first. [Start the benchmark](https://clients.newzealand.ai/benchmark) [Book a free CEO briefing call](https://www.newzealand.ai/contact-us/?service=discovery-call&topic=benchmark#book) --- # The five stages and the six steps | The NZ AI Benchmark Source: https://www.newzealand.ai/benchmark/framework/ [The NZ AI Benchmark](https://www.newzealand.ai/benchmark/) / The framework The framework # The five stages and *the six steps.* Everyone is adopting AI. The winners are adapting. Adoption makes your existing business faster. Adaptation redesigns the business AI makes possible. You need both, and only adaptation changes your competitive position. The NZ AI Benchmark measures where an organisation sits on that path using the five stages and six steps of the AI Playbook, published by AI New Zealand. Maturity is not measured by how many tools you have. It is measured by how much you have been willing to redesign the work itself. [Take the benchmark →](https://clients.newzealand.ai/benchmark) [Read the AI Playbook](https://www.newzealand.ai/ai-playbook/) Five stages ## The five stages. Each stage has a description, a trap, and a sensible place to start. The diagnostic question is deliberately unflattering: pick the row that sounds like a normal Tuesday, not the row you aspire to. Stage 1 ### Aware #### What it looks like Leaders reading headlines and feeling pressure. A few have tried ChatGPT at home. Curiosity, but no permission, no plan, no shared understanding. #### The trap Mistaking awareness for progress. The cost is the quiet one: the gap widens every week you wait. #### Where we start Lunch & Learn, CEO briefing. Stage 2 ### Experimenting #### What it looks like ChatGPT bookmarked, Copilot activated. A few people doing impressive things, most are not. No policy, no standard, no visibility on what works. #### The trap This is the highest risk stage, not the safest. Your best people are feeding company context into ungoverned tools, and the wins live in whoever is using AI that day. #### Where we start Executive AI Briefing, AI Discovery Workshop. Stage 3 ### Adopting #### What it looks like Tools rolled out properly. A policy, an ambassador, training, and solid use cases delivering real time savings. Feels like success. #### The trap The adoption ceiling. You made the legacy way faster without asking whether the legacy way is still right. #### Where we start Use case scoping, AI Sprint, workflow mapping. Stage 4 ### Adapting #### What it looks like You stop asking which tool and start asking how the business should sense, think, decide and act with AI in the loop. Methodology, brand voice and workflows encoded once, connected to real systems, used by everyone, improved monthly. #### The trap Trying to adapt everything at once. Adaptation is operating model surgery. Choose the few workflows where the prize justifies the redesign, and adopt everywhere else. #### Where we start AI Roadmap, AI Operating System. Stage 5 ### Frontier #### What it looks like Populations of agents running adapted workflows. New roles appear: people who own agent quality, set risk appetite, and assure agents in production the way a factory assures its line. #### The trap Forgetting the discipline. Frontier is not a licence to skip governance. Agent assurance is a new budget line, not an optional extra. #### Where we start Retainer, agent assurance advisory. Where most people actually sit: between Experimenting and Adopting, believing they are further along than they are. The next move is always one stage up, not five. The gap ## The Adaptation Gap. The gap is the line between Stage 3 and Stage 4. Between making the old work faster, and redesigning the work itself. 87 percent of New Zealand organisations use some form of AI and only 12 percent have scaled it across the whole business (Datacom State of AI Index 2025). That distance is the Adaptation Gap, and it is where almost every New Zealand organisation is currently stuck. AI does not fix a broken process. It accelerates it. So the strategy question was never which tool. It is how you want the business to run when AI is in the loop. That is not an adoption problem, it is an operating model problem. Governance sits on the commercial side of this ledger too. 62 percent of New Zealanders say they would stop using a product or service over concerns about how the organisation uses AI (One NZ AI Trust Report, May 2026). Clear rules are what let a manager say yes to putting AI in a real workflow, which is why organisations with them tend to move faster, not slower. Six steps ## The six steps, at two altitudes. Six moves take a business from Aware to Adopting. The same six, done deeper, carry it into Adapting and the Frontier. The most common mistake is treating them as a checklist you finish once. They are a discipline you keep levelling up. The benchmark scores how far up each one you are. 1 ### Assign an AI Ambassador. #### Adoption altitude, to Stage 3 One person leads education and adoption, gets tools into hands safely, gives nervous staff their first win. Choose the person who knows your customers and your process. #### Adaptation altitude, Stages 4 to 5 Becomes an AI council and a centre of enablement. Leadership drives it together, aligned to the business plan. Not delegated: federating it out to run independently does not scale. 2 ### Set governance protocols. #### Adoption altitude, to Stage 3 A clear, organisation wide AI policy. A statement of what you will and will not do with AI, so staff can make confident, responsible decisions. #### Adaptation altitude, Stages 4 to 5 Grows into agent assurance. Once agents act in production and talk to other agents, you need a quality system: sampling, audit, and assurance profiles that tighten when an agent drifts. 3 ### Identify practical use cases. #### Adoption altitude, to Stage 3 Workshops to find quick wins. A few repetitive, time consuming tasks to build capability, get value on the board fast, and build belief. #### Adaptation altitude, Stages 4 to 5 Becomes workflow redesign. Map how a critical workflow senses, thinks, decides and acts, then rebuild it around AI rather than bolting AI on. Move a metric, not a use case. 4 ### Understand your risks. #### Adoption altitude, to Stage 3 Stand up an AI committee. Review data, IP, security and privacy against how employees could actually use AI tools. #### Adaptation altitude, Stages 4 to 5 Becomes a leadership discipline: an explicit risk appetite for how close to the frontier you will operate, and owned assurance so problems are caught early. Prompts are not controls. Real controls are technical. 5 ### Select the right AI tools. #### Adoption altitude, to Stage 3 A selection framework prioritising security, scalability and integration with your data. Favour tools that serve multiple use cases. Avoid the shiny object trap. #### Adaptation altitude, Stages 4 to 5 Tools become a system: a shared AI operating system with governance, process, knowledge, integration and skills layers. For some, sovereign AI matters here: models curated for your culture, values and customers. 6 ### Provide ongoing AI skills training. #### Adoption altitude, to Stage 3 Practical, hands on training so the team can use the tools and operationalise new workflows. Power up fast adopters with advanced skills. #### Adaptation altitude, Stages 4 to 5 You build an AI native workforce. Roles deepen, they do not disappear. Frontline staff own agent quality, experts move from firefighting to calibrating, leaders invest in literacy across the whole organisation. Method ## How the benchmark scores. Forty questions, five per dimension, across eight dimensions of how the business runs. Each question offers five descriptions and you pick the one closest to today. Weighted scores roll up into a dimension score and an overall stage. Scoring is rules based and runs on our servers, so the same answers always produce the same result. LeadershipStrategy & Leadership Step 1People & Culture The context engineData & Knowledge Step 5Tools & Systems Step 3Use Cases & Workflows Step 2Governance & Protocols Step 6Skills & Training Step 4Risk, Privacy & Responsible AI Not sure is a valid answer to every question. It is left out of the dimension score and counted as a visibility gap instead, so honesty is never penalised. A dimension with fewer than two answered questions is reported as not assessed rather than scored. Benchmarking happens in two layers. The first compares you with the five stages and best practice, and it is available immediately. The second compares you with other New Zealand organisations of a similar industry, size or region. That layer switches on for a segment once at least 20 organisations in it have completed the benchmark. Until then we say the comparison is not ready rather than show a number built on too few responses. Respondents to the benchmark are not a representative sample of New Zealand business, and every peer view says so. The New Zealand questions draw on the Privacy Act 2020 and Privacy Commissioner guidance, MBIE's Responsible AI Guidance for Businesses, the Algorithm Charter for Aotearoa New Zealand, Te Mana Raraunga principles on Māori data sovereignty, and the AI Forum's AI Blueprint for Aotearoa. Credits ## Who built what. The five stage model, the Adaptation Gap and the six steps are AI New Zealand's published [AI Playbook](https://www.newzealand.ai/ai-playbook/). That is the thinking the benchmark measures against. The scoring engine and the question architecture, including the eight dimensions and the weighted five option scale, were originally developed by [Cairn](https://cairn.nz) and are adapted here with permission. We have recalibrated the stage names, thresholds, commentary and recommendations to the AI Playbook. Credit where it is due: the mechanics are theirs, the position is ours. Free, 20 minutes ## How does your business stack up? Forty questions, your stage and your next move on screen straight away. [Take the benchmark](https://clients.newzealand.ai/benchmark) [Book a free CEO briefing call](https://www.newzealand.ai/contact-us/?service=discovery-call&topic=benchmark#book) --- # AI Lunch n Learn | AI New Zealand Source: https://www.newzealand.ai/lunch-n-learn/ AI Lunch & Learn · 1 hour · online or in-person # Host a Lunch & Learn for the *whole team.* Power up your people with practical knowledge to leverage AI in the workplace. Book your team this engaging session to showcase how AI could supercharge their effectiveness at work. [Book a Lunch n Learn](https://www.newzealand.ai/contact-us/) See what you'll learn **No limit** on attendees **From $950** + GST and travel **Live demos** and open Q&A Who it's for ## Who is this session for? We host Lunch n Learn sessions online and in-person for business teams and community organisations. Businesses kicking-off their AI journey Organisations keen to engage their people in workplace AI Any business who wants their people informed on up-to-date AI know-how, capability and applications ### Terms & conditions. - 👉 There's no limit on the number of attendees - 👉 Attendees must be employed or contracted by the organisation - 👉 Recordings are only for internal use and must never be published or shared externally - 👉 From **$950 plus GST** and travel costs The session ## What you'll learn. This one hour Lunch n Learn will inform, inspire and entertain, showcasing the capability of AI today with time for discussions. Understand what AI is, the types of AI, use cases, capabilities and limitations Practical use cases and real-world applications Establish the ROI for AI. How to identify the uses for AI that will deliver quick wins and long term business value The AI for business tools and platforms and how to choose the right tools for your business Good governance protocols to transparently communicate how the business is using AI and inform decisions on responsible use Risk identification and management to reduce IP leakage and manage data privacy Leadership mindset: tackle common challenges and empower teams for the AI era [Book a Lunch n Learn](https://www.newzealand.ai/contact-us/) Let's start today ## Power up your potential with practical AI skills. Join our community of ambitious leaders and take your career to the next level. [Book an AI Chat](https://www.newzealand.ai/contact-us/?service=ai-chat&topic=lunch-n-learn#book) [Join the Academy](https://academy.newzealand.ai/) --- # Microsoft Copilot Certificate Course | AI New Zealand Source: https://www.newzealand.ai/microsoft-copilot-course/ Certify by AI New Zealand · Certificate course # Become genuinely useful *with Microsoft Copilot.* The **Microsoft Copilot Certificate Course** teaches you to work safely inside your organisation's data, hand real jobs to Copilot in Word, Outlook, Forms, Excel, PowerPoint and Teams, and use Researcher, Analyst and custom agents. One realistic office move, start to finish. [Start the course](https://certify.newzealand.ai/courses/microsoft-copilot) [Training a team?](https://certify.newzealand.ai/courses/microsoft-copilot/team) **$235 + GST** per person, with lower rates for teams. **4** sections, **18** lessons **14** quizzes **Verifiable** certificate About **5 hours**, self-paced ## What you'll be able to do. In the Microsoft 365 apps you already use. - Work safely inside Enterprise Data Protection - Write a prompt that works first time - Hand real jobs to Copilot in six apps - Check the work before it goes anywhere - Brief Researcher and Analyst - Build a custom agent on your own knowledge Needs a paid Microsoft 365 Copilot licence. Why this course exists ## A licence on every desktop isn't adoption. This course addresses the real blockers we've encountered in New Zealand businesses: capability cracks, confidence issues, and the gap between access and meaningful adoption. You close that gap by doing real jobs in Copilot and checking them, not by watching a tool walkthrough. ### Capability cracks. Copilot is switched on, but skills vary wildly across the team. A few power users carry the load while everyone else taps the surface. ### Confidence issues. People worry about trusting the output, breaking something or looking silly. So they quietly go back to the old way of working. ### Access without adoption. A licence doesn't change how work gets done. Adoption takes real jobs handed to Copilot and checked before they go anywhere, which is what every section of this course is built on. Copilot now works inside your files. The skill is knowing what to hand it, what it may touch, and how closely to check what comes back. Who this is for ## Anyone with a Copilot licence and work to do. Anyone in a New Zealand organisation with a Microsoft Copilot licence, whether new to AI or already the person the team asks. ### People new to AI. It starts with what generative AI is, how it predicts rather than knows, and why its output always needs checking. Then a prompting framework that gets a usable answer first time. ### The person the team already asks. Section 3 goes further: research briefs for Researcher, Analyst for what the data says, scheduled prompts, and custom agents built on your own knowledge, with the governed path to sharing them. ### Organisations rolling out Copilot. Word, Outlook, Forms, Excel, PowerPoint and Teams, one real job in each, so people see where Copilot fits their own week rather than in a demo. ### People wary of trusting the output. Much of the course is about checking: the green shield, version history as a safety net, tracing a number back to its source, and spotting when Copilot invents one. ### What it is not. It is not a Microsoft certification, and it is not affiliated with or endorsed by Microsoft. It is not for anyone without a paid Microsoft 365 Copilot licence. If you want the rules before the tools, or have no AI tool yet, start with [AI at Work Foundations](https://www.newzealand.ai/ai-foundations-course/). If your organisation works in Claude, see [Claude at Work](https://www.newzealand.ai/claude-at-work/), and if you are setting AI strategy, the [AI Leaders Course](https://www.newzealand.ai/ai-leaders-course/). How it works ## One office move, all the way through. **Harbourline** is moving office, and Sarah is organising it. Short scenes from Sarah set up each job, and you use Copilot to do it: review the lease safely, turn a lazy prompt into a project plan, draft the staff move guide, write the all-staff announcement, build the staff survey and turn three unmatched quotes into a budget. Then the townhall deck, the issues and actions from the townhall recording, a benchmark of the move against New Zealand practice, the post-move staff sentiment survey, a Harbourline FAQ assistant, and finally a post-move review for the board. Real jobs, in the apps you already use, with the checking built in. On a Business account ### Your own AI Policy, built in. Upload your company AI Policy and Guidelines, and your staff apply the course frameworks and practical exercises to their own work in Copilot, inside your rules rather than a sample company's. - The frameworks are applied to your policy, not a generic one - Every practical exercise lands on their own tasks - Nobody is taught one standard and then handed another [Set up a Business account →](https://certify.newzealand.ai/courses/microsoft-copilot/team) The curriculum ## Four sections, eighteen lessons. After each quiz you see which questions you have right. Try again as many times as you need: each section opens once every quiz in the one before it is passed at 100 percent. 1 ### Copilot Basics. 4 quizzes Understand what generative AI is and where it goes wrong, work inside Enterprise Data Protection with your organisation's data, and write a GCSE prompt that gets a usable answer first time. It closes with the office move's one-page project charter, and the moment Copilot invents a number. - Introducing Generative AI - Safety first - The GCSE framework: Goal, Context, Source and Expectations - Prompting with purpose - Case study: The office move 2 ### Copilot Skills. 6 quizzes Hand a real job to Copilot inside Word, Outlook, Forms, Excel, PowerPoint and Teams, scope what it may touch, and check the work before it goes anywhere. Each app lesson opens with a short scene from Sarah and ends with a piece of the move done. - What has changed - Drafting and Editing (Word) - Communication (Outlook) - Data Collection (Forms) - Analysis and Modelling (Excel) - Visual Storytelling (PowerPoint) - Meeting Management (Teams) 3 ### Copilot Advanced. 4 quizzes Brief Researcher with a CLEAR prompt, get Analyst to tell you what the data says, build a custom agent on your own knowledge, and pull a whole project into an executive report. - Advanced Prompting (CLEAR) - Deep Research (Researcher) - Deep Analysis (Analyst) - Custom Agents - Case Study 4 ### Conclusion and thanks. Free benchmark Wrap up the office move, and see where your organisation sits on the [AI Playbook](https://www.newzealand.ai/ai-playbook/): a free 20 minute benchmark against its five stages. - Benchmark Ready to hand real work to Copilot? [Start the course](https://certify.newzealand.ai/courses/microsoft-copilot) What you leave with ## Finish certified, not just informed. Every exercise is a real job in the Harbourline office move, done in Copilot. By the end, these exist. 1 **A one-page project charter**The lease, the risks and the timeline consolidated in Copilot Chat, and the number Copilot invented spotted. 2 **A staff move guide and an all-staff email**Drafted with Copilot in Word and Outlook, with version history as your safety net and Draft Instructions that teach Copilot how you write. 3 **A staff survey and a budget**A survey built and analysed in Forms, and three unmatched quotes turned into a budget in Excel, with the figures traced back to their source. 4 **A townhall deck and its actions**A deck built from your own documents in PowerPoint, then the issues and actions pulled out of the townhall recording in Teams. 5 **Research, analysis and an agent**A Researcher benchmark of the move against New Zealand practice, an Analyst read of staff sentiment, and the Harbourline FAQ assistant. 6 **A post-move review for the board**Everything you produced, pulled into one executive report using the frameworks you have learned. ### A verifiable certificate, earned on every quiz. To earn it, you correctly answer every quiz and complete every exercise to the required standard. After each quiz you see which questions you have right and can try again as many times as you need, and each section opens once every quiz in the one before it is passed at 100 percent. The certificate is issued by AI New Zealand. It is not a Microsoft certification. From our Copilot training > "I recently attended the Copilot training with New Zealand AI, and it was an excellent experience. Ben was a great facilitator, the session was clear, easy to follow, and provided practical skills we can immediately apply. It's given us a strong foundation to start practising and improving our productivity. I would recommend it for anyone looking to build confidence and efficiency with AI tools." Natalie Chrystall Smith & Smith > "That was really useful. Just what I needed to see Copilot in action through Ben's hands-on and can-do capability and direction." Zac P. Microsoft Copilot Training Before you start ## For yourself, or for your organisation. Individuals and small teams ### Just you, or up to five people. $235 per person + GST - Register and pay by card, then start straight away - Buy up to five seats at once for a small team - A paid Microsoft 365 Copilot licence is needed - Self-paced: about 5 hours, stop between sections and come back [Start the course](https://certify.newzealand.ai/courses/microsoft-copilot) Verifiable certificate issued by AI New Zealand. Vendor-neutral: not affiliated with, or endorsed by, Microsoft or any other AI vendor. Organisations ### Training a team, a workforce or your members. - **Your own course page**, under your name and branding, with white label options for larger accounts - **Upload your company AI Policy and Guidelines** so your staff apply the frameworks and practical exercises to their own work - **Lower rates from the first seat**, and volume pricing that holds when you train in cohorts - **Seats by invitation, one invoice**, and a view of who has completed. Answers stay private to each person [**Business account**6 to 49 people](https://certify.newzealand.ai/courses/microsoft-copilot/team) [**Corporate account**50 or more people](https://certify.newzealand.ai/partner?type=corporate) [**Industry association account**Offer it to your members, with a member discount or a revenue share](https://certify.newzealand.ai/partner?type=association) Not sure which fits? [Talk to us](https://www.newzealand.ai/contact-us/). Good to know ## Frequently asked questions. **Do I need a Copilot licence?** Yes, a paid Microsoft 365 Copilot licence. The course has you working with Copilot inside Word, Outlook, Forms, Excel, PowerPoint and Teams, and we would rather you knew that now than found out halfway through a section. **I'm new to AI. Is it for me?** Yes. It is written for anyone with a Copilot licence, whether new to AI or already the person the team asks. Section 1 starts with what generative AI is and why its output always needs checking. **How do I earn the certificate?** Correctly answer every quiz and complete every exercise to the required standard. There are 14 quizzes across the first three sections, and each section opens once every quiz in the one before it is passed at 100 percent. **What if I get a quiz question wrong?** After each quiz you see which questions you have right. Try again as many times as you need. **Is it a Microsoft certification?** No. The certificate is issued by AI New Zealand. It is not affiliated with any Microsoft training courses or certifications, and does not constitute Microsoft endorsed training. **How is it different from AI at Work Foundations?** [AI at Work Foundations](https://www.newzealand.ai/ai-foundations-course/) is tool-neutral: the rules, the risks and the judgement before anyone is handed a tool. This course is hands-on with Microsoft Copilot, for people who already have a licence. **Will my employer see my answers?** No. On a Business account your organisation sees who has completed and the outcome. Your answers stay private to you. **How do I buy it for my team?** Up to five seats can go on a card. From six to 49 people, [set up a Business account](https://certify.newzealand.ai/courses/microsoft-copilot/team), invoiced at the lower rate from the first seat. From 50, [apply for a Corporate account](https://certify.newzealand.ai/partner?type=corporate). Associations can [offer it to their members](https://certify.newzealand.ai/partner?type=association). The AI Academy ## Keep up with AI, *without keeping up with all of it.* Join the AI Academy for free and get our weekly newsletter on the most useful applications of AI for business. [Join the Academy →](https://academy.newzealand.ai/) --- # Claude at Work Certificate Course | AI New Zealand Source: https://www.newzealand.ai/claude-at-work/ Certify by AI New Zealand · Certificate course # From prompts to projects. *Claude that knows your work.* The **Claude at Work** course takes you past ad hoc chats to an AI that knows your organisation, writes in your voice, follows your standards, and takes whole jobs off your desk while you keep the judgement. Practised on your own real work. [Start the course](https://certify.newzealand.ai/courses/claude-at-work) [Training a team?](https://certify.newzealand.ai/courses/claude-at-work/team) **$235 + GST** per person, with lower rates for teams. **4** sections, **17** lessons **9** quizzes **Verifiable** certificate About **3 hours**, self-paced ## What you'll be able to do. On your own work, not sample data. - Use Claude safely with your organisation's information - Write prompts that get usable work first time - Set up a Project that knows your organisation - Produce documents that come out on brand - Turn one of your workflows into a reusable Skill - Schedule work that is done before you start the day Needs a paid Claude plan with Cowork access. Why this course exists ## Better screenshots, same bottlenecks. That is AI theatre, and most organisations are quietly stuck in it. A few people do genuinely impressive things while everyone else saves four minutes a day and wonders what the fuss was about. This course is the step from asking Claude questions to handing it whole jobs. ### You are the memory. Every new conversation starts with you explaining the organisation again. Who we are, what we sell, how we write, what we never do. Same paragraph, every time, because the context lives in your head instead of in the tool. Section 2 puts it in a Claude Project. ### It looks like it came from nowhere. The content is good and the document is wrong. Wrong colours, wrong fonts, no logo, and fifteen minutes of fixing headings before it can go anywhere near a client. A brand reference file and a styled starter template fix it once. ### The expertise is walking around. One person has run the thing for years and none of it is written down. It is not that they are difficult. Documenting is a writing job, and it loses to real work every time. The course captures it for about an hour of their time, and they correct rather than write. The difference is almost never talent. It is that nothing the good ones know has been written down where the tool can reach it. Who this is for ## Knowledge workers with a paid Claude plan. People in New Zealand organisations who have Claude and want it doing real work: documents, data, research and the recurring jobs that eat the week. ### Daily users who have plateaued. Section 1 will be quick for you, and it is meant to be. The rest is built around the parts most daily users have never set up: organisational memory, brand control, workflow redesign, Skills and agent guardrails. ### People who keep re-explaining their organisation. Build a Claude Project from a small, curated set of files about how your organisation works, so it knows your context before you type a word. ### People who own a recurring process. Map one of your own, decide which steps an agent owns and which stay with a person, name the metric you will move, then turn it into a Skill that runs the same way every time. ### Organisations putting Claude to work. Each person writes your organisation's three-line rule for confidential information, learns what an agent can reach and where it must stop and ask, and leaves with a plan for taking it wider across the organisation. ### What it is not. It is not a first introduction to AI at work: for the rules, the risks and what stays your job, start with [AI at Work Foundations](https://www.newzealand.ai/ai-foundations-course/). If Microsoft 365 is your main environment, the [Microsoft Copilot course](https://www.newzealand.ai/microsoft-copilot-course/) is the hands-on equivalent. Leaders setting AI strategy want the [AI Leaders Course](https://www.newzealand.ai/ai-leaders-course/). And it is not for the free Claude plan: Section 3 runs on Cowork, which needs Pro, Max, Team or Enterprise. How it works ## Your own work, start to finish. Four sections, seventeen lessons, about 3 hours, self-paced. Section 1 gets you using Claude safely and prompting with purpose, and ends on a case study: turn a 41-page request for proposal into a one-page decision brief for leadership, then check the claims that would be most expensive to get wrong. From there the exercises run on one folder of your own real work, which you pick before you start. You build a Project, a brand file, a redesigned workflow and a Skill, and in Section 3 Claude runs multi-step jobs across real files. The Claude desktop app, on macOS or Windows, is recommended for that section. Not sample data. Something you actually deal with, so you finish with something you can use on Monday. On a Business account ### Your own AI Policy, built in. Upload your company AI Policy and Guidelines, and your staff apply the course frameworks and practical exercises to their own work in Claude, inside your rules rather than generic ones. - The frameworks are applied to your policy, not a generic one - Every practical exercise lands on their own work - Nobody is taught one standard and then handed another [Set up a Business account →](https://certify.newzealand.ai/courses/claude-at-work/team) The curriculum ## Four sections, seventeen lessons. From first prompt to work that runs before you start the day. The course is linear: after each quiz you see which questions you have right, you try again as many times as you need, and each section opens once every quiz in the one before it is passed at 100 percent. 1 ### Foundations: platform, prompts and purpose. 5 lessons · 2 quizzes Use Claude safely with your organisation's information, write prompts that get specific, usable work first time, and know which part of Claude each job belongs in. - Meet Claude - Safety first - Prompting with purpose - Chat, Projects, Artifacts and Cowork - Case study: read the tender 2 ### Real work: memory, brand, data, evidence. 5 lessons · 3 quizzes Set up a Claude Project that knows your organisation, produce on-brand documents, analyse data you can check, research with sources you can defend, and capture expert know-how without asking anyone to write it down. - Projects and organisational memory - On-brand documents - Working with data - Research and evidence - Capturing what is in someone's head 3 ### Agentic: Cowork, Skills, guardrails. 5 lessons · 3 quizzes Brief Claude to complete multi-step jobs across real files, redesign one of your own workflows, turn it into a reusable Skill, limit what an agent can reach, and schedule work that is done before you start the day. - Cowork: from chat to task - Redesign one workflow - Build a Skill - Guardrails and agent security - Scheduled tasks and connectors 4 ### Wrap up. 2 lessons · 1 quiz Leave with a four-week plan for putting what you built into daily use, and your certificate. - Your first 30 days with Claude - Your certificate and what's next Stop briefing your AI from zero every morning. [Start the course](https://certify.newzealand.ai/courses/claude-at-work) What you leave with ## More than a certificate. The ability to delegate Claude workflows that free up your time, and the frameworks to reset how you work with AI in the loop. Built on your own work, so by the end these exist. 1 **A Claude Project that knows your organisation**Built from a small, curated set of files about how your organisation works, so it has your context before you type a word. 2 **A brand file and starter template**So documents come out looking like your organisation the first time, without you fixing headings afterwards. 3 **One workflow redesigned**One of your own recurring processes, mapped: the steps an agent owns, the steps that stay with a person, and the metric you will move. 4 **A reusable Skill**Your redesigned process, running the same way every time, whoever is at the keyboard. 5 **A scheduled task**Real work done before you open your laptop, written to stand on its own and never send anything without you. 6 **A four-week plan**Week by week, for putting your Project, brand file, workflow and Skill into daily use, and taking it wider across your organisation. ### Finish certified, not just informed. To earn the certificate you answer each quiz correctly and complete every exercise to the required standard. After each quiz you see which questions you have right and can try again as many times as you need; each section opens once every quiz in the one before it is passed at 100 percent. The result is a verifiable certificate issued by AI New Zealand. Before you start ## For yourself, or for your organisation. Individuals and small teams ### Just you, or up to five people. $235 per person + GST - Register and pay by card, then start straight away - Buy up to five seats at once for a small team - You need a paid Claude plan with Cowork access (Pro, Max, Team or Enterprise) and one folder of your own real work - Self-paced: about 3 hours, stop between sections and come back [Start the course](https://certify.newzealand.ai/courses/claude-at-work) Verifiable certificate issued by AI New Zealand. Vendor-neutral: not affiliated with, or endorsed by, Anthropic or any other AI vendor. Organisations ### Training a team, a workforce or your members. - **Your own course page**, under your name and branding, with white label options for larger accounts - **Upload your company AI Policy and Guidelines** so your staff apply the frameworks and practical exercises to their own work - **Lower rates from the first seat**, and volume pricing that holds when you train in cohorts - **Seats by invitation, one invoice**, and a view of who has completed. Answers stay private to each person [**Business account**6 to 49 people](https://certify.newzealand.ai/courses/claude-at-work/team) [**Corporate account**50 or more people](https://certify.newzealand.ai/partner?type=corporate) [**Industry association account**Offer it to your members, with a member discount or a revenue share](https://certify.newzealand.ai/partner?type=association) Not sure which fits? [Talk to us](https://www.newzealand.ai/contact-us/). Good to know ## Frequently asked questions. **What do I need before I start?** A paid Claude plan with Cowork access: Pro, Max, Team or Enterprise. The free plan will not get you through Section 3, and we would rather you knew that now. The Claude desktop app on macOS or Windows is recommended for Section 3. And one folder of your own real work to practise on: not sample data, something you actually deal with. Pick it before you start. **Do I need to be technical?** No. It is written for knowledge workers, not developers, and the exercises run on everyday work: documents, data, research and a recurring process of your own. **I already use Claude every day. Is this too basic?** Section 1 will be quick for you, and it is meant to be. The course is built around the parts most daily users have never set up: organisational memory, brand control, workflow redesign, Skills, agent guardrails and scheduled tasks. If you have already built a Project with a curated knowledge base and a Skill that runs a recurring job, you are ahead of it. **Does this work if my organisation uses Copilot as well?** Yes. The organisational memory you build in Section 2 is a curated set of files about how your organisation works, so it goes with you to whatever tool you point at it. If Microsoft 365 is your main environment, look at the [Microsoft Copilot course](https://www.newzealand.ai/microsoft-copilot-course/) too. **How long does it take?** About 3 hours, self-paced, ideally over two or three sittings rather than one. You can stop between sections and come back. **What if I get a quiz question wrong?** Try again, as many times as you need. After each quiz you see which questions you have right, and each section opens once every quiz in the one before it is passed at 100 percent. **Is it an accredited qualification?** No. It is a verifiable certificate issued by AI New Zealand, earned by answering every quiz correctly and completing every exercise to the required standard. It is not affiliated with, or endorsed by, Anthropic, and is not Anthropic endorsed training. **Will my employer see my answers?** No. On a Business account your organisation sees who has completed and the outcome. Your answers stay private to you. **How do I buy it for my team?** Up to five seats can go on a card. From six to 49 people, [set up a Business account](https://certify.newzealand.ai/courses/claude-at-work/team), invoiced at the lower rate from the first seat. From 50, [apply for a Corporate account](https://certify.newzealand.ai/partner?type=corporate). Associations can [offer it to their members](https://certify.newzealand.ai/partner?type=association). The AI Academy ## Keep up with AI, *without keeping up with all of it.* Join the AI Academy for free and get our weekly newsletter on the most useful applications of AI for business. [Join the Academy →](https://academy.newzealand.ai/) --- # The AI Leaders Course | AI New Zealand Source: https://www.newzealand.ai/ai-leaders-course/ Certify by AI New Zealand · Certificate course # Make AI safe to say yes to. *The AI Leaders course.* Either nothing has been approved and everyone is waiting, or everyone has quietly got on with it and the risks are starting to show. Both land on the same desk. Yours. [Start the course](https://certify.newzealand.ai/courses/ai-leaders) [Training a leadership team?](https://certify.newzealand.ai/courses/ai-leaders/team) **$495 + GST** per person, $295 a seat for five or more. **7** sections, **38** lessons **7** practical exercises **Human-marked** assessment About **10.5 hours**, self-paced ## What you'll be able to do. Six things you can't do today. - Get your AI mandate in writing - Know which AI tools staff use - Tier risk and approve tools - Pick the AI ideas worth doing - Cost AI and show what it returned - Defend AI decisions to the board No paid AI subscription and no software to buy. Who this is for ## The remit is bigger than the title. You might have been given a title. You might have just been nodded at in a leadership meeting. Either way the job reaches a long way past whatever was said out loud. You are deciding what the organisation is allowed to do, what it buys, how its people work, where its risk sits, and what it tells its clients when they ask. That is an operating model, not a tool rollout. Nobody trains for it, and whoever builds it shapes how the place works for years. ### Operations, finance and people leaders. GMs, practice leads, operations managers, finance and people directors, and the people who became the de facto AI person. You run things for a living. AI is now one of the things you run. ### Organisations of 50 to 500 people. New Zealand organisations big enough that scattered adoption is a real risk, where responsibility for AI has landed on a leader who now has to face a board. ### People who have used ChatGPT or Copilot. That is the whole prerequisite. No code, no prompt engineering curriculum, no model architecture. This course teaches judgement, which is the part nobody has written down for you. ### Anyone who has to face a board. Or an executive team, a partnership, a tender panel or an insurer. If someone can ask you to explain and defend how your organisation uses AI, this gets you ready for the question. ### We built it around four fears. Because they are the four we hear every week, from capable people who are doing this on top of a full job. - Looking naive in front of the board - Approving something that causes a privacy breach - Spending money you cannot justify - Being the person who slowed everything down, or the person who let it run wild The job is not to be the AI expert. It is to be the person who makes good AI use happen safely, and can show their working. ### Who it is not for. If what you want is to get better at using the tools yourself, this is the wrong course: the [Claude at Work](https://www.newzealand.ai/claude-at-work/) or [Microsoft Copilot](https://www.newzealand.ai/microsoft-copilot-course/) course is the right one, and either pairs well with this in either order. If your staff need the basics before they are given access, that is [AI at Work Foundations](https://www.newzealand.ai/ai-foundations-course/). And if you want model architecture, agent frameworks or prompt engineering, we do not teach any of them here. Why this course exists ## Widespread, and shallow. Almost every New Zealand organisation has started. Almost none has finished. Individuals are getting real value from these tools and the organisation is getting almost none of it, while carrying all of the risk. Breadth · who is using AI78% Depth · strategic integration12% Source: 2026 [NZ AI Roadshow](https://www.newzealand.ai/roadshow/) Attendee Report. ### Nobody wrote down who decides. Your MD said "we need to be all over this" at a leadership meeting and meant it. Nothing was written down and nobody was given authority, so months later you are still asking permission for the things you were supposedly put in charge of. ### It is already happening without you. People are using AI at work today and not telling you. Your data is not leaking. It is being handed over, by good people, in the absence of anything better provisioned, and a better policy will not stop the concealment. ### You cannot cost it or prove it. Hours saved with no destination is not a benefit, and any chair worth having will say so. No baseline, no approval. And without a measured column sitting next to the estimated one, what you have is a slide deck. How it works ## You are not studying a case. You have the job. From the first lesson you are the AI Lead at **Calder Rowe**, a New Zealand engineering, planning and environmental consultancy. They are not a technology company. Neither, probably, are you. You answer to an independent Chair whose most-used sentence is "says who?", a Managing Director who will say yes to anything in a meeting, a Finance and Operations Director who holds the budget, and an IT Manager who has been carrying the risk alone. The work compounds. The register you build in Section 4 is the register you cost in Section 6, and every piece is yours to defend in front of the Chair. Bring your own organisation's numbers as you go: headcount, licence spend, and what IT can tell you is already in use. On a Business account ### Your own AI Policy, built in. Upload your company AI Policy and Guidelines, and your people apply the course frameworks and practical exercises to their own organisation, inside your rules rather than Calder Rowe's. - The frameworks are applied to your policy, not a generic one - Every practical exercise lands on your own organisation - Nobody is taught one standard and then handed another [Set up a Business account →](https://certify.newzealand.ai/courses/ai-leaders/team) The curriculum ## Seven sections, thirty eight lessons. Every section ends in a quiz and a practical exercise. After each quiz you see which questions you have right, and you can try again as many times as you need. Each section opens once every quiz in the one before it is passed at 100 percent. 1 ### The Mandate. 1 quiz · 1 practical exercise The difference between the body that decides and the body that diffuses. Which of four maturity stages your organisation is in, and which of the four AI OS layers you have. Then a one-paragraph mandate with decision rights, budget, escalation list, reporting cadence and an end date, signed. And three questions to interrogate any statistic. - The job is not what you think it is - Where your organisation actually sits - A mandate with teeth - Says who? 2 ### The Honest Baseline. 1 quiz · 1 practical exercise Why prohibition produces less safety than provision. A three-part baseline: tool discovery, a five-question anonymous pulse and an honest governance inventory. Why a better policy will not stop concealment, the two questions that matter to ask vendors about your data, and how to tell a notifiable privacy breach from an IT ticket. - Shadow AI is a provisioning failure - Your before picture - Why people hide it - Three things to say out loud - When it is a breach, not an IT ticket 3 ### The People. 1 quiz · 1 practical exercise The people-first argument, made as a commercial argument. What a generative AI tool does, in plain English, and why it is confidently wrong. The Four Gates in five seconds, the Delegation Grid for any task, and the Three Checkpoints where the human stays. Then the line between space to play and the bar for proposing, and your first three use cases. - The case you have to make - A shared mental model of how these tools work - The Four Gates: what data do I share? - The Delegation Grid: what can I hand over? - The Three Checkpoints: where does the human stay? - From understanding to real work 4 ### The Register. 1 quiz · 1 practical exercise Walk a real workflow and find where value leaks. A ten-field intake that kills about a third of submissions on purpose, and a register with eight status values. Adapted RICE with a risk multiplier, and what the score does not know. Now, Next, Later and Not doing, published. And how to tell a knowledge problem from an AI one. - Use cases are the wrong unit - The intake form and the register - Scoring without pretending - Five things the score will not catch - Now, Next, Later, Not doing - The Org Brain 5 ### The Guardrails. 1 quiz · 1 practical exercise Which existing New Zealand laws already bite on AI use. A risk tier assigned in under five minutes, and a tool assessed against six sections and seven critical items. An incident run from report to systemic change. One firm-wide answer for a client, tender panel or insurer. And who may build an agent, and who owns it in six months. - The legal floor in New Zealand - Tier the intake or strangle yourself - Approving tools - Guidelines first, policy second - When it goes wrong - The client conversation - Who may build an agent 6 ### The Money. 1 quiz · 1 practical exercise A defensible twelve-month AI cost position that keeps actual, estimated and unknown apart. How to read a token-based invoice, four cost controls to set before a pilot, and the value measured and governance health sections of a quarterly board report. - No baseline, no approval - Where the money actually goes - Tokens, in plain English - Four controls that cost nothing - The one page the board actually reads 7 ### Keeping It Alive. 1 quiz · 1 practical exercise An AI Champions Network chosen by nomination, including one person who is not convinced. Why recognition is a control, not a perk. A dashboard of six to eight numbers that separates leading from lagging. Then how to read a stall, and the recovery plan. - Building the Network - Recognition is the control, not the perk - Measuring what actually changed - It will stall - Before the final assessment ### The final assessment. **Your work, defended in front of a Chair who asks "says who?".** The mandate, the register, the tier model, the cost position and the ninety day plan for Calder Rowe are each written by you. Marked by an AI New Zealand assessor against a [published standard](https://certify.newzealand.ai/courses/ai-leaders/rubric) you can read before you start. **A person marks it, not a machine.** Stop asking permission for the thing you were put in charge of. [Start the course](https://certify.newzealand.ai/courses/ai-leaders) What you leave with ## More than a certificate. Ten and a half hours to build the structure that fixes it. Clear authority to decide, an honest picture of what your people are already doing, guardrails that speed approvals up, a shortlist of the work worth doing, a budget you can defend, and a ninety day plan with names and dates on it. 1 **A mandate**One paragraph with decision rights, budget, escalation list, reporting cadence and an end date, written to be signed. 2 **A use case register**Fed by a ten-field intake, scored with adapted RICE and a risk multiplier, and published as Now, Next, Later and Not doing. 3 **A tier model**Risk tiers you can assign in under five minutes, and a way to approve tools without becoming the bottleneck. 4 **A cost position**Twelve months, with actual, estimated and unknown on separate lines, and four cost controls set before any pilot. 5 **A ninety day plan**Written by you, and defended in front of a board chair who asks "says who?". 6 **The certificate**AI Leaders Course, issued by AI New Zealand on a human-marked final assessment. ### Certified on assessed work, not attendance. Answer every quiz correctly, complete every exercise to the required standard, then pass the final assessment, marked by an AI New Zealand assessor against a [published standard](https://certify.newzealand.ai/courses/ai-leaders/rubric). The certificate says you can do the work, not that you sat through it. Before you start ## For yourself, or for your organisation. Individuals and small teams ### Just you, or a small team. $495 per person + GST - Register and pay by card, then start straight away - Five or more people pay $295 a seat - No paid AI subscription and no software to buy - Self-paced: about ten and a half hours, ideally across three or four weeks [Start the course](https://certify.newzealand.ai/courses/ai-leaders) Certificate issued by AI New Zealand after human assessment. Vendor-neutral: not affiliated with, or endorsed by, any AI vendor. Organisations ### Training a leadership team, a workforce or your members. - **Your own course page**, under your name and branding, with white label options for larger accounts - **Upload your company AI Policy and Guidelines** so your people apply the frameworks and practical exercises to their own work - **$295 a seat from the first seat** on a Business account - **Seats by invitation, one invoice**, and a view of who has completed. Answers stay private to each person [**Business account**6 to 49 people](https://certify.newzealand.ai/courses/ai-leaders/team) [**Corporate account**50 or more people](https://certify.newzealand.ai/partner?type=corporate) [**Industry association account**Offer it to your members, with a member discount or a revenue share](https://certify.newzealand.ai/partner?type=association) Not sure which fits? [Talk to us](https://www.newzealand.ai/contact-us/). Good to know ## Frequently asked questions. **Do I need to be technical?** No. There is no code in this course and no prompt engineering. If you have used ChatGPT or Copilot, that is the whole prerequisite, and if you can read a board paper and form a view, you can do all of it. **Do I need a paid AI tool?** No. There is no paid AI subscription and no software to buy. It is useful to bring your own organisation's numbers: headcount, licence spend, and what IT can tell you is already in use. **Is this the same as the Claude at Work or Microsoft Copilot courses?** No, and they solve different problems. Those two teach you to use a tool well. This one teaches you to lead the organisation's use of all of them, and it is deliberately vendor-neutral. Doing a tool course as well is a good idea, in either order. See the [Claude at Work course](https://www.newzealand.ai/claude-at-work/) and the [Microsoft Copilot course](https://www.newzealand.ai/microsoft-copilot-course/). **How is it different from AI at Work Foundations?** [AI at Work Foundations](https://www.newzealand.ai/ai-foundations-course/) is for the person about to be handed the tools: the rules, the risks and what stays their job. This course is for the person accountable for AI across the organisation: the mandate, the register, the guardrails and the costs. **Is this an accredited qualification?** No. It is a certificate issued by AI New Zealand, resting on a final assessment marked by a person against a published standard. We are explicit about what that is and is not worth. **Who marks the final assessment?** An AI New Zealand assessor, against a [published standard](https://certify.newzealand.ai/courses/ai-leaders/rubric) you can read before you start. **What if I get a quiz question wrong?** After each quiz you see which questions you have right. Try again as many times as you need. Each section opens once every quiz in the one before it is passed at 100 percent. **Do you offer group rates?** Yes. Five or more people pay $295 a seat plus GST. For six to 49 people, [set up a Business account](https://certify.newzealand.ai/courses/ai-leaders/team), which pays $295 from the first seat. From 50, [apply for a Corporate account](https://certify.newzealand.ai/partner?type=corporate). Associations can [offer it to their members](https://certify.newzealand.ai/partner?type=association). **Can I share my access with a colleague?** No. It is an individual enrolment. If several people need it, five or more pay $295 a seat, and a Business account pays that rate from the first seat. The AI Academy ## Keep up with AI, *without keeping up with all of it.* Join the AI Academy for free and get our weekly newsletter on the most useful applications of AI for business. [Join the Academy →](https://academy.newzealand.ai/) --- # AI at Work Foundations Certificate Course | AI New Zealand Source: https://www.newzealand.ai/ai-foundations-course/ Certify by AI New Zealand · Certificate course # What you should know *before working with AI.* The **AI at Work Foundations course** covers the rules you work inside, what these tools are and where they fail, what they can already reach, and what stays your job. Written for New Zealand workplaces, with a final assessment marked by a person. [Start the course](https://certify.newzealand.ai/courses/ai-at-work-foundations) [Training a team?](https://certify.newzealand.ai/courses/ai-at-work-foundations/team) **$235 + GST** per person, with lower rates for teams. **11** sections, **46** lessons **5** practical exercises **Human-marked** assessment About **4 hours**, self-paced ## What you'll be able to do. Before anyone hands you a licence. - Know what you may put into an AI tool - See what your tools can already reach - Spot a confident wrong answer - Decide what to hand over, and what to keep - Check the work before your name goes on it - Ask for a tool an approver can say yes to No AI tool, licence or subscription needed to take it. Why this course exists ## "Can they use AI?" is the wrong question. The question an employer is actually asking is: am I confident this person will use our AI tools responsibly? This course is the knowledge an organisation wants its people to have before it gives them access, and a certificate that says they have it. ### "AI skills" has no definition yet. Job ads ask for it and CVs claim it, with nothing shared underneath. Most training fills the gap with prompt tips and tool tutorials, which date in months and miss the part that matters. ### The guardrails come before the tools. What you may put in, what a tool can reach without you pasting anything, and who to ask when you are not sure. Section 2 of this course, before anyone is taught to get more out of a tool. ### The human sent it. It was not the AI that sent the wrong figure. It was the person who did not check it. Accountability does not transfer, and a signature is not a review. The course does not prove somebody is good at AI. It proves they know the rules, the risks, and where their own judgement has to stay. Who this is for ## Anyone about to be handed AI tools at work. And anyone who wants to show an employer they will use them responsibly. Any role, any industry, any level of confidence. ### Teams starting an AI programme. The people who will get a licence in the next few months, and whose employer needs to know it is safe to hand it over. Step one, before the tools are switched on and before any tool training. ### New starters and people changing role. Useful from day one: three questions to ask, a standard you have written for yourself, and a tool request you could actually send. ### People applying for roles that ask for AI skills. A defensible answer when the job ad asks, without a degree or a technical background. Every practical exercise works on the role you are aiming at as well as one you hold. ### People who are wary of AI. Much of the course is about what stays human, what not to hand over, and how to say no with a reason. It is written to work for the person who thinks this is coming for their job. ### What it is not. It is not tool training: there is no click-by-click on Copilot, ChatGPT, Claude or Gemini. That is what the [Claude at Work](https://www.newzealand.ai/claude-at-work/) and [Microsoft Copilot](https://www.newzealand.ai/microsoft-copilot-course/) courses are for, and they come after this one. It is not for leaders setting AI strategy, who want the [AI Leaders Course](https://www.newzealand.ai/ai-leaders-course/), and it is not technical: no code, no model architecture, no building agents. How it works ## One company, all the way through. **Tasman Fresh** is a Nelson food and beverage distributor of 62 people, supplying cafés, restaurants and grocers across the top of the South Island. It buys from a Māori owned orchard and supplies two iwi owned hospitality groups. Microsoft 365, no AI policy, and most of what it knows living in people's heads. Every section opens with something happening there, beginning on the first Tuesday in March and resolving a quarter later. You watch the company change as you go. Then you do each practical exercise on your own work, in the job you have or the one you are going for. Not on Tasman Fresh's. On a Business account ### Your own AI Policy, built in. Upload your company AI Policy and Guidelines, and your staff apply the course frameworks and practical exercises to their own work, inside your rules rather than a sample company's. - The frameworks are applied to your policy, not a generic one - Every practical exercise lands on their own tasks - Nobody is taught one standard and then handed another [Set up a Business account →](https://certify.newzealand.ai/courses/ai-at-work-foundations/team) The curriculum ## Eleven sections, forty six lessons. The course is linear and nothing is optional. Each section ends in a short quiz, and the next one opens once every answer is right. Retake as often as you need: the feedback explains the principle, not the answer. 1 ### Where the work actually goes. 1 quiz See where your working day actually goes, and why making one step faster changes little unless it is the step where the work waits. - The day that disappears - Knowledge stuck in heads - The copy and paste tax - Faster is not the same as better 2 ### The rules you work inside. 1 quiz · 1 practical exercise What you may put into a tool, what it can already reach, where your obligations come from, and who to ask when you are not sure. Inside those lines you can move without asking. - What AI guidelines are for - The Four Gates - Commercial risk is not personal information - Three things that stay true whatever tool you use - Whose permissions is it using? - When a gate stops you - Where the rules actually come from 3 ### What you are actually working with. 1 quiz · 1 practical exercise Explain what an AI tool is and how it fails, using three images you can hold, drawing on Dr Karaitiana Taiuru's Kaupapa Māori AI Framework. And why your knowledge of your own work stays the final check. - A person, a puppet, a shadow - Whose hand is on the strings - A shadow has no light of its own - Where it fails, and why it agrees with you - You are the subject matter expert 4 ### What the tools can reach. 1 quiz The kinds of AI you will meet, what ChatGPT, Copilot, Claude and Gemini can already see, who owns what they produce, and which tool and account to use for work. - The kinds of AI you will meet - The four you will actually use - Who owns the output - The notetaker in the room - Which tool, and which account 5 ### What you put in, and what you sign off. 1 quiz · 1 practical exercise Decide what to hand to a tool and what to keep, check what comes back against the source, and name who is accountable before anything is sent. - The Delegation Grid - Verification at source - The Three Checkpoints - It was not the AI that sent it 6 ### When AI does things, not just drafts them. 1 quiz Tell a tool that drafts for you from one that acts on its own, and know the one rule and four questions that keep any of them safe near real records. - Ask and act, or brief and done - Prompts, skills, agents and plain old automation 7 ### The Org Brain. 1 quiz Why a tool's answers depend on what your organisation has written down, and the question to ask of any task: what would this need to know? - Four layers of knowledge - Layer three is a folder, not a technology project - What would this need to know? 8 ### Context is the job. 1 quiz Give a tool the context only you have, argue with its first draft, and make sure what you work out does not stay with you. - The stranger and the friend - GCSE: Goal, Context, Source, Expectations - Sense, think, decide, act - Argue with it - What you worked out should not stay with you 9 ### Cultural sensitivity: whose knowledge is this? 1 quiz · 1 practical exercise Recognise material you hold but have no right to use, including Māori data and knowledge, and know who should ask and who decides. Te Tiriti o Waitangi taught from its own text. - Where ownership is not the test - Māori data and taonga - The relationship, not the compliance box - What te Tiriti asks 10 ### The value you bring. 1 quiz Separate the work you can hand over from the judgement that is your job, see what these tools can give back to people working around a barrier, and face honestly what happens to work that goes. - Delegate the repetition, keep the judgement - Critical thinking is the moat - The tool that removes a barrier - Roles deepen. Some tasks do not survive. 11 ### What to ask before you start. 1 quiz · 1 practical exercise The three questions that show how ready your organisation is, how to ask for a tool in a way an approver can act on, and what to do and say when something looks wrong. - Your first three questions - The Tool Case - When something looks wrong, and what you tell people ### The final assessment. **Three workplace scenarios, and you choose the one closest to your own work.** Each asks the same five questions: your call, the framework, the sensitive judgement, where it happens in your own work, and where you would push back. Answered in writing, in about an hour, with no clock. You may use AI tools; the answers must be about your own work. Marked by a named AI New Zealand assessor against a [published standard](https://certify.newzealand.ai/courses/ai-at-work-foundations/rubric) you read before you start. **Judgement, reasoning and honesty score. Fluency does not.** Know the rules before you are handed the tools. [Start the course](https://certify.newzealand.ai/courses/ai-at-work-foundations) What you leave with ## More than a certificate. Five practical exercises, each done in your own role or the one you are aiming at. By the end, these exist. 1 **Your own AI standard**Written down: what you will not put in, how you decide when you are not sure, and who you ask. The standard you work to until someone writes one for you. 2 **Five tasks on the Delegation Grid**Five of your own tasks, sorted by whether you could undo the result and whether you could check it. 3 **A six sentence explanation**What an AI tool is and how it fails, in your own words, tested on somebody else. 4 **Three questions for your manager**Where the source of truth is, where use is registered, and where problems get reported. Booked, not just noted. 5 **A tool request you could send tomorrow**Written as the job to be done rather than a product, so an approver can act on it. 6 **The certificate**AI at Work Foundations, issued by AI New Zealand on a human-marked final assessment. ### Certified on judgement, not attendance. Answer every quiz correctly, complete every exercise to the required standard, then pass the final assessment, marked by an AI New Zealand assessor against a [published standard](https://certify.newzealand.ai/courses/ai-at-work-foundations/rubric). It certifies knowledge and judgement before access. It does not claim to prove skill with any particular tool, and we say so. Before you start ## For yourself, or for your organisation. Individuals and small teams ### Just you, or up to five people. $235 per person + GST - Register and pay by card, then start straight away - Buy up to five seats at once for a small team - No AI tool, licence or subscription needed - Self-paced: about four hours, stop between sections and come back [Start the course](https://certify.newzealand.ai/courses/ai-at-work-foundations) Certificate issued by AI New Zealand after human assessment. Vendor-neutral: not affiliated with, or endorsed by, any AI vendor. Organisations ### Training a team, a workforce or your members. - **Your own course page**, under your name and branding, with white label options for larger accounts - **Upload your company AI Policy and Guidelines** so your staff apply the frameworks and practical exercises to their own work - **Lower rates from the first seat**, and volume pricing that holds when you train in cohorts - **Seats by invitation, one invoice**, and a view of who has completed. Answers stay private to each person [**Business account**6 to 49 people](https://certify.newzealand.ai/courses/ai-at-work-foundations/team) [**Corporate account**50 or more people](https://certify.newzealand.ai/partner?type=corporate) [**Industry association account**Offer it to your members, with a member discount or a revenue share](https://certify.newzealand.ai/partner?type=association) Not sure which fits? [Talk to us](https://www.newzealand.ai/contact-us/). Good to know ## Frequently asked questions. **Do I need an AI tool or a licence to take it?** No. Nothing in the course needs a paid AI tool, a licence or a policy. It is written for the point before you are given one. **Will it teach me to use Copilot, ChatGPT or Claude?** No, deliberately. It is tool-neutral, so it stays true when the tools change. For hands-on skills, the [Claude at Work](https://www.newzealand.ai/claude-at-work/) and [Microsoft Copilot](https://www.newzealand.ai/microsoft-copilot-course/) courses come next. **How is it different from the AI Leaders Course?** This one is for the person doing the work. The [AI Leaders Course](https://www.newzealand.ai/ai-leaders-course/) is for the person accountable for AI across the organisation: the mandate, the register, the guardrails and the costs. **Is it an accredited qualification?** No. It is a certificate issued by AI New Zealand, resting on a written assessment marked by a person against a published standard. It certifies knowledge and judgement before access, not skill with a particular tool. **Can I use AI in the final assessment?** Yes. The answers must be about your own work, and what scores is judgement, reasoning and honesty rather than fluency. The [assessment standard](https://certify.newzealand.ai/courses/ai-at-work-foundations/rubric) is published so you can read it first. **What if I get a quiz question wrong?** Try again, as many times as you need. The feedback explains the principle the question tested rather than naming the right option, so a retake still needs the reasoning. Each section opens once every quiz in the one before it is passed. **Will my employer see my answers?** No. On a Business account your organisation sees who has completed and the outcome. Your answers stay private to you. **Whose work does it draw on?** Section 3 uses Dr Karaitiana Taiuru's Kaupapa Māori AI Framework, He Tangata, He Karetao, He Ātārangi, and section 9 draws on his position that Māori data is taonga, with his permission and attributed in the lessons. See [taiuru.co.nz](https://www.taiuru.co.nz/kaupapa-maori-ai-framework/). **How do I buy it for my team?** Up to five seats can go on a card. From six to 49 people, [set up a Business account](https://certify.newzealand.ai/courses/ai-at-work-foundations/team), invoiced at the lower rate from the first seat. From 50, [apply for a Corporate account](https://certify.newzealand.ai/partner?type=corporate). Associations can [offer it to their members](https://certify.newzealand.ai/partner?type=association). The AI Academy ## Keep up with AI, *without keeping up with all of it.* Join the AI Academy for free and get our weekly newsletter on the most useful applications of AI for business. [Join the Academy →](https://academy.newzealand.ai/) --- # The Great NZ AI Roadshow | AI New Zealand Source: https://www.newzealand.ai/roadshow/ The 2026 Great NZ AI Roadshow # The Great NZ AI Roadshow is for Kiwis striving to *work smarter, move faster and grow.* We hosted 8 events across Aotearoa during May & June. Attendees discovered how to shift from ad-hoc AI use, where AI occasionally benefits a few. To a trusted system that enables the whole company to flourish. By the numbers ## The 2026 Roadshow wrap up. 1,400+ Attendees 500+ Businesses 96.8% Rated the day useful or better The ethos is simple ## Super-practical advice to help your business leverage AI thoughtfully and responsibly. In 2026 AI New Zealand brought the best local know-how and practical expertise to businesses and organisations around Aotearoa to advance their AI adoption. ### Business Owners & Directors. Your competitors are already using AI to work faster, win more jobs, and protect their margins - this day shows you exactly how to do the same in a practical way. You'll leave with a simple action plan you can start on Monday, built around real examples from businesses that look like yours. ### People & Culture Leaders. AI is already reshaping roles across Aotearoa, and this day gives you the language, frameworks, and confidence to lead that change instead of reacting to it. You'll walk away with ready-to-use resources you can plug straight into your internal training. ### Business & Technology Leaders. You're under pressure to "do something about AI", and this day gives you the clear, evidence-based view you need to decide what's right for your organisation. You'll see what comparable NZ businesses are actually doing, how they're managing risk, and how to evaluate AI investments against real outcomes like revenue, cost, and customer experience. ### Business Services Professionals. You're already experimenting with AI, and this day shows you how to turn those experiments into reliable, professional-grade workflows that save you serious time. You'll see live demos, steal proven workflows from experts and peers, and leave knowing exactly which tools to double down on for your role. Attendee feedback ## Actual attendee feedback from our Auckland event. Don't just take our word for it, here's what Roadshow attendees submitted in the post event survey. 8/10 > "From start to finish, the roadshow was highly informative, non-condescending and engaging! It covered all aspects of where to start for smaller/medium businesses with minimal IT support and next to zero current AI usage. A very inspiring day leaving us feeling energised and confident." 7/10 > "All the speakers were really engaging in their allocated timeslots (which were also a perfect length to get across their topic without becoming tedious)." > "Great mix of speakers, and fast pace was good. Really enjoyed the creative AI segment, where AI was used for some less serious endeavours (that can have seriously good outcomes)." > "Small breakout sessions or workshops where business specific questions can be asked. Happy to pay more for this type of event." 6/10 > "You needed to explain the jargon. Tokens and Agents. No idea but met someone at the event who explained it to me." **** Roadshow Partners ## The biggest initiative to boost AI adoption in NZ. Brought to you with the 2026 Great New Zealand AI Roadshow Partners. The programme ## What the 2026 Roadshow covered. ### The leadership culture for successful AI adoption. What if the real AI advantage isn't the tools you buy, but the way your leaders ask questions, make decisions, and turn "AI theatre" into how work actually gets done? ### Practical advice to prepare data & infrastructure for AI. If AI is the engine, your data and infrastructure are the fuel lines. Are you really ready to plug AI into your business, or would it just accelerate outdated processes? ### How to manage data privacy, risk & compliance. When AI can see everything, who decides what it should never touch? How sure are you that your current privacy and risk settings aren't quietly undermining trust? ### Why a hybrid workforce (AI+Humans) is essential. When AI agents can run tasks end-to-end, the real advantage isn't replacing people - it's how you redesign work so agents handle the repetition and your humans double down on high-stakes judgement, creativity and relationships. ### The workflows you can optimise with AI today. If you picked just one workflow this quarter, one bottleneck, one metric - what would change if AI prepared the work and your team focused purely on high-value judgement? ### Get clarity on the skills and tools your business needs. Build a plan to get started or accelerate by focusing on the skills & tools that will unlock capability & capacity to deliver value more effectively. Where we went ## Eight cities across Aotearoa. May and June 2026. ### Hamilton. Claudelands Thanks to everyone who joined us in Hamilton. SOLD OUT ### Wellington. Tākina Thanks to everyone who joined us in Wellington. SOLD OUT ### Christchurch. Town Hall Thanks to everyone who joined us in Christchurch. SOLD OUT ### Auckland. Shed 10 Thanks to everyone who joined us in Auckland. SOLD OUT ### Nelson. Rutherford Hotel Thanks to everyone who joined us in Nelson. Completed ### Napier. Convention Centre Thanks to everyone who joined us in Napier. Completed ### New Plymouth. International Hotel Thanks to everyone who joined us in New Plymouth. SOLD OUT ### Tauranga. The Cargo Shed Thanks to everyone who joined us in Tauranga. SOLD OUT The speakers ## Who presented at the 2026 Roadshow. ### Justin Flitter Founder of NewZealand.AI Since founding AI New Zealand in 2017 Justin has become one of Aotearoa's leading AI for business advisors. In 2018 & 2019 he produced the AI-DAY conferences. In 2024 Justin hosted the AI Roadshow series with AWS. He's spoken at over 200 business and industry conferences and worked with 100's of businesses to advance their AI enablement programs with practical advice & training. Justin represented New Zealand on the Global Partnership for AI and is a Director of Business Mentors New Zealand. ### Richard Kenyon Associate Director of AI at Datacom Richard leads the AI strategy practice at Datacom, advising clients on AI adoption and governance. ### Anchali Anandanayagam Partner at Hudson Gavin Martin Anchali leads the HGM's technology, media and IP practice and advises on AI-related development, procurement and commercialisation. She is recognised globally by Chambers & Partners, Legal 500 and Media Law International for her expertise in AI, emerging technologies and data privacy law in New Zealand. ### Asa Cox Founder of Arcanum Asa Cox is the Founder and CEO of Arcanum AI, a Wellington-based company pioneering accessible AI solutions for SMBs across New Zealand and Australia. Asa presented at the 2024 Roadshow and will facilitate the practical workshop. We're delighted to welcome Asa back on the Roadshow bus. ### Sarah Trotman, ONZM CEO of Business Mentors NZ As Chief Executive of Business Mentors NZ, Sarah is a driving force behind the integration of AI into BMNZ's business as usual. Sarah will share their journey, from developing an AI Mentoring Assistant and a Digital Mentor to implementation and adoption of those initiatives that are now transforming small business productivity. BMNZ, a not for profit, has matched an experienced, volunteer business mentor to 85,000+ small business owners. ### Raymond Schippers AI & Security Expert Raymond Schippers is a cybersecurity executive and AI security strategist with over 15 years of experience building enterprise-wide programs for global organisations. As Lead Technologist for AI & Security at Check Point Software, he partners with government and industry leaders to design secure-by-design strategies for AI adoption and defence. Previously at Canva, he scaled the Detection & Response group, built 24/7 global SOC and CTI teams, and pioneered the company's strategy for detecting and defending against AI-enabled threats. ### Larissa Logan Founder of ALYRT Larissa Logan is a Licensed Insolvency Practitioner and Fellow Chartered Accountant with 20+ years' experience advising businesses through growth, stress and turnaround. She is the Founder of ALYRT, an AI-powered platform helping SMEs gain real-time visibility over cash, risk and performance, starting with construction, translating complex data into clear actions as the advisor in your pocket. ### Andrew Dentice Partner at Hudson Gavin Martin Andrew specialises in technology, data and commercial law, advising clients of all sizes on technology transactions, data protection, commercialisation and regulatory issues, especially in fintech, crypto, SaaS and platform business models. He regularly presents on AI matters, advises on AI transactions and regulation, and has led HGM's internal AI policies and deployment. ### Edwin Lim Partner at Hudson Gavin Martin Ed brings nearly 25 years of experience in the technology, media and telecommunications sector, enabling him to provide comprehensive technology advice to organisations ranging from large enterprises developing major systems to start-ups entering the market. ### Simon Martin Founding Partner at Hudson Gavin Martin Simon has been immersed in the technology, media and IP sector since 1990, guiding companies through business operations, growth and purpose while securing the rights they need. He regularly advises clients of all sizes on AI-related issues such as licensing AI tools, joint ventures to develop tools and compliance, using his deep sector insight to build deals that meet their specific needs effectively. ### Steve Ballantyne Brand strategist at Brand IQ Steve Ballantyne is a brand strategist, storyteller, AI expert, and founder of Brand IQ. He helps founders and marketers use AI-powered storytelling to sharpen positioning, create standout content, and scale ideas faster. With 30+ years in marketing, he also teaches AI-powered storytelling at the New Zealand Marketing Association and Australian Marketing Institute. ### Ben Gleisner Founder & CEO at Cogo Ben Gleisner is an economist, environmentalist, and social entrepreneur. A Victoria University of Wellington graduate, Ben founded Cogo because he believes in the power of people, technology and data to fundamentally change the world for good. Under his leadership, Cogo has grown from a New Zealand startup into a global climate technology company, partnering with major financial institutions across four continents to drive measurable climate impact at scale. ### Sang Yun Byun Cybersecurity at Check Point Sales Engineer at Check Point, specializing in cybersecurity and the evolving threat landscape. His work focuses on helping organisations defend against modern attacks, including emerging risks driven by AI, while exploring practical approaches to securing AI systems themselves. ### Eva Hartshorn-Sanders Director of Hartsheba Eva Hartshorn-Sanders is a global specialist in online safety and technology policy, and the director of Hartsheba. She has worked for more than 20 years in policy, law reform and politics, which includes high profile work with the United Nations, NZ Government, in the UK Parliament, and for national and international clients. Eva's experience includes briefing the White House, members of the US Congress, UK Parliament, and the EU on emerging research and policy on online safety. ### Matt Crowther Co-founder, CentaurAI Matt co-founded CentaurAI to help New Zealand individuals and businesses turn AI curiosity into measurable performance. Based in Hawke's Bay, CentaurAI builds team capability and governance through its progressive training programmes and is a trusted source of guidance for businesses navigating AI, across primary industries, professional services and trades. With 30+ years of operations leadership and governance experience across construction, manufacturing and primary industries, Matt brings a practical and research backed lens to AI adoption. ### Kieran Lee Founder of N16 Consulting Kieran Lee is the founder of N16 Consulting, a Taranaki-based advisory firm helping New Zealand SMEs adopt AI and automation in ways that make a genuine operational difference. With 20+ years of experience in operational improvement across global and NZ businesses, and credentials in Lean Six Sigma and Human-Centred Service Design, Kieran takes a business-first approach to AI adoption - diagnosing the problem before reaching for the tool. N16 is an approved provider under the Regional Business Partner Network's (MBIE) AI Advisory Pilot, delivering AI roadmaps and implementation support to SMEs across New Zealand. ### Serge van Dam Speaker · Wellington Roadshow Serge presents "AI for Boards and Directors" at the Wellington Roadshow event. ### Wendy Folau-Kennedy Founder, Attune · Creator of Pattern Intelligence™ Wendy Folau-Kennedy is a transformation strategist, educator, futurist and creator of Pattern Intelligence™. She helps Boards, CEOs and business owners read the human patterns shaping AI adoption, the patterns that decide whether technology lands or fails. She brings 120,000+ coaching hours across elite sport, Fortune 500 boards, government and high-growth business. Her recent Substack piece *Provenance Is Not the Point* drew direct responses from Fonterra's Chief AI Officer and from senior commentators in the audit world. Let's start today ## Power up your potential with practical AI skills. Join our community of ambitious leaders and take your career to the next level. [Book an AI Chat](https://www.newzealand.ai/contact-us/?service=ai-chat&topic=roadshow#book) [Join the Academy](https://academy.newzealand.ai/) --- # About AI New Zealand | AI New Zealand Source: https://www.newzealand.ai/about-us/ About us # We've been accelerating AI adoption in Aotearoa *since 2017.* Justin Flitter founded AI New Zealand in 2017 to help business leaders discover why, where and how to leverage artificial intelligence. Since then we have helped thousands of Kiwi organisations with practical advice and training. We are independent and vendor neutral. We do not sell software licences, so the advice you get is about what will work for your organisation rather than what we have to move. [**LinkedIn** AI New Zealand](https://www.linkedin.com/company/ai-new-zealand/) [**YouTube** @newzealand_ai](https://www.youtube.com/@newzealand_ai) [**Instagram** @newzealand_ai](https://www.instagram.com/newzealand_ai/) [**Email** hello@newzealand.ai](mailto:hello@newzealand.ai) What we have built ## Nine years of AI in New Zealand. ### New Zealand AI Academy. Provides skills based training at scale to individuals, business teams and industry groups. Free and paid webinars and workshops are hosted every week to share practical advice, demonstrate tools and tech and showcase the practical and remarkable ways AI is changing the way we work. ### Industry Partnerships. We're proud to work with forward thinking, future focused industry organisations. Together we're providing the learning and development programs their members need to confidently leverage Artificial Intelligence. ### Keynote Presentations. Justin Flitter has presented at 100's of industry and company conferences. His expertise and enthusiasm make him one of New Zealand's most sought-after keynote speakers on AI. He brings his signature energetic style and practical advice to help people supercharge their potential with AI. ### The Great NZ AI Roadshow. The 2024 New Zealand AI Roadshow was attended by 1,800 people across 10 cities. These super-practical, AI for business masterclass events are perfect for non-technical people learning why, where & how to leverage AI at work. ### AI Day Conference. AI Day was New Zealand's headline artificial intelligence conference, held in both 2018 and 2019, for business leaders, and everyone interested in the impact of AI on society and business. Both events were seen as foundational moments for New Zealand's AI ecosystem, helping foster confidence, accelerate adoption and grow the community. ### The AI Show. The AI Show was one of the country's most influential AI event series from 2017 through 2024. These monthly meet-ups, attended by up to 300 people were held in Auckland and consistently attracted business leaders, technical professionals, and people curious about the practical impacts of AI. Our founder ## Justin Flitter. [Connect on LinkedIn](https://www.linkedin.com/in/justinflitter/) Justin Flitter is the founder of AI New Zealand, the country's independent AI for business advisory. He has spent his career at the forefront of emerging technology, building the communities and events that shaped New Zealand's AI ecosystem, and now advises boards and executive teams on why, where and how to use AI. He graduated from Massey University in 1999 with a Bachelor of Business Information, majoring in marketing and communications management. Before AI New Zealand he led the marketing programs for high-growth technology and professional services businesses and was the seventh hire at Zendesk, leading business development across Asia Pacific. He has also founded or led nine business communities with more than 75,000 combined members. He's represented New Zealand on the Global Partnership on Artificial Intelligence, helps produce TEDxAuckland, and is a Director on the Board of Business Mentors New Zealand. Justin is based in Auckland. #### Advise. Independent strategy, governance and roadmaps for boards and executive teams. #### Teach. Practical, skills based training through workshops and the AI Academy, including Microsoft Copilot for M365, the AI Ambassador programme and live webinars. #### Build. The AI Operating System, one governed, trusted AI environment your whole team can work in, plus AI agents and custom development. #### Share. The weekly AI Toolkit, The Great New Zealand AI Roadshow, and commentary in the national media. Career history | Years | Role | | --- | --- | | 2017 – present | Founder, AI New Zealand and the New Zealand AI Academy | | 2022 – present | Director and Board Member, Business Mentors New Zealand | | 2012 – present | Producer, TEDxAuckland | | 2021 – 2026 | Chairperson, NZ Tech Marketers Group | | 2017 – 2024 | Founder, The AI Show and AI-Day | | 2009 – 2012 | Co-founder, Auckland Social Media Club | | 2008 – 2011 | Business Development APAC, Zendesk (seventh hire) | | 2006 – 2008 | Founder, Wellington Young Professionals Group | | 2000 – 2004 | Founder, NRG Trust Business Incubator | - **700+ speaking events** delivered across industry and company conferences. - **Nine communities** founded or led, with more than 75,000 combined members. - **Global Partnership on Artificial Intelligence**, representing New Zealand. - **TVNZ NZ Marketing Awards** judge, 2019 to 2023. Work with us ## Let's talk about where AI fits in your organisation. Book a 15 minute AI chat, meet the team, or start with the AI Playbook. [Contact us](https://www.newzealand.ai/contact-us/) [Read the AI Playbook](https://www.newzealand.ai/ai-playbook/) --- # Contact AI New Zealand | AI New Zealand Source: https://www.newzealand.ai/contact-us/ Let's talk AI # Contact *Us.* Book a 20 minute AI Chat or a 45 minute Discovery Call with Justin Flitter or Ben Walker to plan AI enablement in your organisation. Pick a time below and it goes straight into our calendars. Book an AI Chat Book a Discovery Call [**Email** hello@newzealand.ai](mailto:hello@newzealand.ai) [**Instagram** @newzealand_ai](https://www.instagram.com/newzealand_ai/) [**LinkedIn** AI New Zealand](https://www.linkedin.com/company/ai-new-zealand/) [**YouTube** @newzealand_ai](https://www.youtube.com/@newzealand_ai) Book a time ## Pick a time that suits you. A 20 minute AI Chat, a 45 minute Discovery Call or a 60 minute Planning Session, online with Justin or Ben. Choose a time and we'll send a calendar invite with a Teams link. [20 minutes **AI Chat** A quick call about our services, a use case, an event, or anything else. Email us three times that suit you.](mailto:hello@newzealand.ai?subject=Book%20an%20AI%20Chat&body=Hi%20team%2C%0A%0AI%27d%20like%20to%20book%20a%2020%20minute%20AI%20Chat.%20Three%20times%20that%20suit%20me%3A%0A%0A) [45 minutes **Discovery Call** A focused conversation about AI strategy, team training or your first AI-powered workflows. Email us three times that suit you.](mailto:hello@newzealand.ai?subject=Book%20a%20Discovery%20Call&body=Hi%20team%2C%0A%0AI%27d%20like%20to%20book%20a%2045%20minute%20Discovery%20Call.%20Three%20times%20that%20suit%20me%3A%0A%0A) [60 minutes **Planning Session** Identify opportunities, prioritise initiatives and build a roadmap that works for your business. Email us three times that suit you.](mailto:hello@newzealand.ai?subject=Book%20a%20Planning%20Session&body=Hi%20team%2C%0A%0AI%27d%20like%20to%20book%20a%20Planning%20Session.%20Three%20times%20that%20suit%20me%3A%0A%0A) Or email [hello@newzealand.ai](mailto:hello@newzealand.ai) and we'll sort a time. Request our services ## See what we do, and what it costs. Our services page covers the AI Operating System, leadership and governance, process and workflow mapping, training, and a typical first engagement, with prices. Tell us who you are and we will send you a private link. We read every request. Your link lasts seven days, and you can share it with colleagues at your organisation. Work email please. Nothing is sent until we approve your request. Our impact ## 9 years and counting. We started ahead of the curve. Now we help organisations across Aotearoa leverage this rapidly changing technology to work smarter, move faster and grow. Advice at the leadership table. Training for the people doing the work. And implementation that embeds AI into everyday workflows. We're proud to work with construction, manufacturing, engineering, financial services, not-for-profit and government organisations, from teams of five to workforces of 500 plus. Client verified metrics 484 Client engagements 495 Use cases evaluated 48,383 Annual hours saved $6.84M Annual costs saved Updated September 2026 · verified from the AI New Zealand client platform. Hours and cost saved are counted from live use cases only. The people ## Meet Our Growing Team. Our team brings together deep expertise across AI strategy, governance, technology architecture, development, training and change management. Every engagement has a named lead, and we bring in the right specialist rather than sending whoever is free. ### Justin Flitter Founder & CEO · Auckland Since founding AI New Zealand in 2017, Justin Flitter has led the partnerships that grew the community & produce the events that informed, inspired & showcased the capability of AI for business & society. Today, with a talented team across Aotearoa, they're a leading provider of skills based training and the implementation of managed and measured AI systems. Justin is also a Director, on the board of Business Mentors New Zealand. [Connect on LinkedIn →](http://linkedin.com/in/justinflitter) ### Ben Walker CTO & Head of Strategy · Dunedin With over 25 years in technology, Ben possesses a deep understanding of technology architecture and the practical applications of Artificial Intelligence. As CTO and Head of Strategy he sets our technical direction and leads our Microsoft Copilot and ChatGPT workshops and training delivery. Ben is a Microsoft Certified AI Business Professional and AI Transformation Leader, and equally comfortable in a tenant configuration conversation or in front of a room of people who have never opened an AI tool. [Connect on LinkedIn →](https://www.linkedin.com/in/benwalkernz/) ### Tiina Carryer AI Advisor & Trainer · Auckland Tiina has been working in product, marketing and ecommerce for over 20 years, holding senior roles at Microsoft across the United Kingdom, Europe and ANZ. She combines global experience with hands-on start-up work in New Zealand, helping businesses turn ideas into practical outcomes. Today, she helps teams build confidence in adopting AI, delivering workshops that make AI accessible and grounded in everyday work. [Connect on LinkedIn →](https://www.linkedin.com/in/tiina-carryer/) ### Steve Wilson Senior Process and AI Analyst · Taranaki Steve Wilson helps New Zealand businesses and councils cut through the AI noise and focus on what actually creates value. 20+ years in business analysis means he asks 'what problem does this solve?' before 'which tool should we use?' He turns what a business says it wants into workflows, levers and a sequenced plan, then stays on to help put it in place. Steve also runs Zero to AI, an education platform for professionals getting to grips with AI, and Ministry of Insights for high-stakes consulting work. He covers Taranaki and the central North Island. [Connect on LinkedIn →](https://www.linkedin.com/in/changeable/) ### James Oldham Lead AI Engineer · Auckland James Oldham builds advanced voice agents, automation workflows and enterprise-grade AI systems for clients across the globe. He designs and develops custom enterprise AI solutions for businesses, ranging from advanced Voice AI solutions and workflow automations to data-driven operational tools that help organisations scale efficiently. He is the person who turns a mapped workflow into something that actually runs, and the one who will tell you early if an idea is harder to build than it looks. [Connect on LinkedIn →](https://www.linkedin.com/in/james2000/) ### Tony Baker AI Engineering & Advisory · Auckland Tony brings more than 26 years’ experience in commercial software and applies that experience to practical AI. He helps New Zealand organisations identify valuable AI opportunities, turn them into delivery plans that fit real-world constraints and build the capability to sustain adoption. Most recently, Tony was a Distinguished Engineer at Yes Energy, where he led AI adoption with forecasting teams across APAC and the United States. Before that, he established and led TESLA Forecasting’s Asia-Pacific business. He holds a Bachelor of Computing and Mathematical Sciences with First Class Honours from the University of Waikato. [Connect on LinkedIn →](https://www.linkedin.com/in/tony-baker-nz/) ### Peter Mangin AI Advisor & Implementation · Auckland Peter Mangin has been working in AI and technology for over 25 years and is our technical expert on tools and AI models. He advises on technical solutions and AI model development, facilitates workshops, and stays across what each platform can actually do rather than what its marketing claims. He has helped more than 400 businesses use AI to find new ways to grow and improve. Peter knows how to turn AI ideas into real-life solutions that make a difference. [Connect on LinkedIn →](https://www.linkedin.com/in/pmangin/) ### Wendy Folau-Kennedy Leadership & Capability Coach · Christchurch Wendy Folau-Kennedy is a leading transformational coach, facilitator and trusted advisor with over 30 years of hands-on experience helping individuals and organisations thrive through disruptive change. Founder of Folau Futures, Wendy partners with leaders across New Zealand, Australia and the globe to deliver bespoke coaching and innovative learning experiences that build resilience, adaptability and future-fit strategy. She works on the part of AI adoption that has nothing to do with technology and everything to do with people. [Connect on LinkedIn →](https://www.linkedin.com/in/wendy-f-679333/) ### Join our team. Our nationwide team of business and technical experts enables us to scope, design, build and deploy AI solutions with local capability. We're always keen to connect with people that share our vision and bring additional capability for our clients. Connect with Justin to start the conversation. [Connect with Justin →](http://linkedin.com/in/justinflitter) Our values ## Business to the power of *AI.* ### Foster Curiosity. The previously unimaginable is now possible. We'll inspire and inform businesses and communities to better understand this remarkable new technology. ### Unlock Potential. We'll showcase how people can turn their curiosity into new AI-powered skills and tools that help New Zealand businesses thrive. ### Be Trusted Partners. We'll deliver independent advice and support to help business leaders make informed decisions, to maximise their potential with AI, every step of the way. Let's start today ## Power up your potential with practical AI skills. Join our community of ambitious leaders and take your career to the next level. Book an AI Chat [Join the Academy](https://academy.newzealand.ai/) --- # Premium Membership | AI New Zealand Source: https://www.newzealand.ai/premium/ AI Academy · Premium Membership # Premium Membership For *Individuals.* For just $59 per month, or $545 a year, you'll gain anytime access to the following AI Academy benefits. [Unlock My Premium Membership](https://academy.newzealand.ai/checkout/premium) See what's included **$59** per month **$545** per year **7-day** free trial **Cancel** anytime What's included ## Your Premium Membership benefits. ### Weekly AI Toolkit Webinars. Perfect for self-motivated professionals keen to learn practical AI skills. - 30-45min sessions (usually) at 11am Fridays - Shared learning and discussions - Business adoption best practices - AI workflow walk-throughs - Discover new AI apps - Keep on top of the latest new features - Monthly or annual subscription, 7-days free trial, cancel anytime ### On-Demand Channel. Missed a webinar or workshop? Catch up in your own time. - Exclusive access to recordings of our best webinars and workshops - Work through the library to get up to speed - Includes our best free and paid events ### Exclusive Invites. Be the first to get course access and event invitations. - Exclusive invites to member-only events - Early bird prices for in-person events - $50 off every certificate course on [Certify](https://certify.newzealand.ai/courses) - Book a free 60-min AI Sprint for your team [Unlock My Premium Membership](https://academy.newzealand.ai/checkout/premium) Included with Premium ## Next in the AI Toolkit. A new practical session every Friday, live in the Academy. Right now we're running the Workflow Sprint: pick three real workflows, build them together across the tools, then deploy them for real. FRI 25 Sep · 11:00am NZT · 45 min ### Build Workflow Two: The Project Activity Tracker Justin Flitter · Workflow Sprint part 2 [Register →](https://academy.newzealand.ai/c/the-ai-toolkit-webinars/workflow-series-2) FRI 9 Oct · 11:00am NZT · 45 min ### Build Workflow Three, Together Justin Flitter · Workflow Sprint part 3 [Register →](https://academy.newzealand.ai/c/the-ai-toolkit-webinars/workflow-series-3) FRI 16 Oct · 11:00am NZT · 45 min ### Deploy All Three, For Real Justin Flitter · Workflow Sprint finale [Register →](https://academy.newzealand.ai/c/the-ai-toolkit-webinars/workflow-series-4) [Unlock My Premium Membership](https://academy.newzealand.ai/checkout/premium) [See all webinars →](https://academy.newzealand.ai/c/the-ai-toolkit-webinars) Real leaders, real impact ## From breakthroughs, to 'aha' moments. > "In just half an hour a week you've boosted my AI confidence 10 fold!" Lisa S. AI Toolkit Webinars > "I found it really useful, particularly understanding the AI models and how they can all be integrated into the strategy." Rob E. AI Ambassador Course > "I've been to a few AI events...but none with takeaways this useful." Amanda G. Cultivate - Women in Finance event Good to know ## Frequently asked questions. **How does the member discount on certificate courses work?** Premium members get $50 off each of our [certificate courses on Certify](https://certify.newzealand.ai/courses), including GST. It applies automatically when you register on Certify with the same email address as your Academy membership, so there is no code to enter. You can use it once per course in any 12 months, on as many different courses as you like. It does not combine with a coupon code: you get whichever saving is larger. **Can I cancel my subscription?** You can cancel anytime. Canceling means you'll lose access to the associated Academy benefits. **Can I get a receipt?** A receipt will automatically be sent to you. If you need it resent, please email [hello@newzealand.ai](mailto:hello@newzealand.ai) **Can I share my account with a colleague?** No. This is an individual membership for you. **Can I download event recordings?** No. We don't have the option to allow video downloads. They must be watched in the Academy. The AI Academy ## Keep up with AI, *without keeping up with all of it.* Join the AI Academy for free and get our weekly newsletter on the most useful applications of AI for business. [Join the Academy →](https://academy.newzealand.ai/) --- # AI Insights and Research for New Zealand Business Source: https://www.newzealand.ai/insights/ Insights # Research and field notes on *AI that actually gets used.* What we are seeing inside New Zealand businesses: what works, what stalls, and why the gap between buying AI and getting value from it is an operating problem rather than a technology one. Every article ## The full archive. Client case story ### [A virtual property manager, built into Aspire's own portal.](https://www.newzealand.ai/aspire-property-management-case-story/) Justin Flitter · October 2026 Read article → AI Operating System ### [Everyone is buying the skills layer. Almost nobody is buying the integration layer.](https://www.newzealand.ai/integration-layer/) Justin Flitter · 3 August 2026 Read article → 2026 AI Roadshow Notes ### [AI is no longer just about content generation. It is becoming operational infrastructure.](https://www.newzealand.ai/operational-infrastructure/) Justin Flitter · 26 June 2026 Read article → Insights ### [What comes after LLMs now has a New Zealand face on it.](https://www.newzealand.ai/what-comes-after-llms-now-has-a-new-zealand-face-on-it/) Justin Flitter · 22 June 2026 Read article → Guidelines ### [You Don't Have to Rent Your AI](https://www.newzealand.ai/you-don-t-have-to-rent-your-ai/) Justin Flitter · 17 June 2026 Read article → Guidelines ### [A New Zealand read on Anthropic's Mythos, the RBNZ warning, and Trump's frontier model order](https://www.newzealand.ai/a-new-zealand-read-on-anthropic-s-mythos-the-rbnz/) Justin Flitter · 9 June 2026 Read article → Guidelines ### [Five things we have learnt so far on the Great New Zealand AI Roadshow.](https://www.newzealand.ai/five-things-we-have-learnt-so-far-on-the-great-new/) Justin Flitter · 7 June 2026 Read article → Guidelines ### [Your People Are Ready for AI. Your Infrastructure Isn't.](https://www.newzealand.ai/your-people-are-ready-for-ai-your-infrastructure-isn-t/) Justin Flitter · 19 March 2026 Read article → Guidelines ### [AI platforms are releasing new functionality that's changing the way we work.](https://www.newzealand.ai/ai-platforms-are-releasing-new-functionality-that-s/) Justin Flitter · 15 March 2026 Read article → Basics ### [When AI Teaches Itself: Why Digital Minds Learn Differently From Humans.](https://www.newzealand.ai/when-ai-teaches-itself-why-digital-minds-learn/) Justin Flitter · 8 March 2026 Read article → Guidelines ### [The Master Prompt to generate a basic AI Policy.](https://www.newzealand.ai/the-master-prompt-to-generate-a-basic-ai-policy/) Justin Flitter · 25 February 2026 Read article → Best Practice ### [The deadsimple AI upgrade most teams miss](https://www.newzealand.ai/the-deadsimple-ai-upgrade-most-teams-miss/) Justin Flitter · 24 February 2026 Read article → Tools ### [The OpenClaw moment and what it means for NZ Business](https://www.newzealand.ai/the-openclaw-moment-and-what-it-means-for-nz-business/) Justin Flitter · 22 February 2026 Read article → Guidelines ### [AI Glossary: 100+ Terms Every NZ Business Leader Should Know.](https://www.newzealand.ai/ai-glossary/) Justin Flitter · 3 February 2026 Read article → Guidelines ### [How to Choose AI Tools for Your Business: A 2026 Comparison Framework for New Zealand Leaders](https://www.newzealand.ai/how-to-choose-ai-tools-for-your-business/) Justin Flitter · 3 February 2026 Read article → Guidelines ### [CEO Briefing: What is Clawdbot (now Moltbot), is it safe to use at work?](https://www.newzealand.ai/ceo-briefing-what-is-clawdbot-now-moltbot-is-it-safe/) Justin Flitter · 2 February 2026 Read article → Guidelines ### [Stop Dabbling: The AI Muscle Gap holding back New Zealand.](https://www.newzealand.ai/stop-dabbling-the-ai-muscle-gap-holding-back-new/) Justin Flitter · 28 January 2026 Read article → Best Practice ### [Capability is cheap. Judgment is rare.](https://www.newzealand.ai/capability-is-cheap-judgment-is-rare/) Justin Flitter · 20 January 2026 Read article → Guidelines ### [The Year AI Stops Being Theatre.](https://www.newzealand.ai/the-year-ai-stops-being-theatre/) Justin Flitter · 15 January 2026 Read article → Critical Thinking ### [AI gave us the answer. Critical Thinking saved the deal.](https://www.newzealand.ai/ai-answer/) Justin Flitter · January 2026 Read article → Value Levers ### [AI use cases are a distraction. Here's what we did instead.](https://www.newzealand.ai/use-cases-distraction/) Justin Flitter · January 2026 Read article → Operating Logic ### [AI doesn't fix a broken process. It just accelerates it.](https://www.newzealand.ai/broken-process/) Justin Flitter · January 2026 Read article → Leadership ### [From AI Theatre to Real Impact: A NZ Financial Services Story.](https://www.newzealand.ai/ai-theatre/) Justin Flitter · January 2026 Read article → Insights ### [AI in Aotearoa in 2025 - By the numbers](https://www.newzealand.ai/ai-in-aotearoa-in-2025-by-the-numbers/) Justin Flitter · 15 December 2025 Read article → Guidelines ### [Essential playbook of 60 AI task-boosters](https://www.newzealand.ai/essential-playbook-of-60-ai-task-boosters/) Justin Flitter · 15 December 2025 Read article → Insights ### [This is the sort of AI initiative New Zealand should be renowned for.](https://www.newzealand.ai/this-is-the-sort-of-ai-initiative-new-zealand-should/) Justin Flitter · 30 November 2025 Read article → CEO Briefing ### [The CEO Briefing: The 2026 AI Imperative.](https://www.newzealand.ai/2026-ceo-ai-briefing/) Justin Flitter · 28 November 2025 Read article → Guidelines ### [AI for Business: Why Now is the Time for Kiwi SMEs to Harness the Power of Artificial Intelligence](https://www.newzealand.ai/ai-for-business-why-now-is-the-time-for-kiwi-smes-to/) Justin Flitter · 16 November 2025 Read article → People & Culture ### [AI for Business: Why the tech needs people and culture to thrive.](https://www.newzealand.ai/ai-needs-people-culture/) Justin Flitter · 11 November 2025 Read article → Guidelines ### [Why AI Models Hallucinate and How to Prevent It.](https://www.newzealand.ai/why-ai-models-hallucinate-and-how-to-prevent-it/) Justin Flitter · 16 October 2025 Read article → AI Agents ### [CEO Briefing: AI Agents Are Doing Actual Work.](https://www.newzealand.ai/ceo-briefing-ai-agents-are-doing-actual-work/) Justin Flitter · 30 September 2025 Read article → Industry Insights ### [New Zealand's AI Adoption: A Comprehensive Analysis in 2025.](https://www.newzealand.ai/new-zealand-ai-adoption-in-2025/) Justin Flitter · 29 September 2025 Read article → Insights ### [The mindset shift for graphic designers in the AI era.](https://www.newzealand.ai/the-mindset-shift-for-graphic-designers-in-the-ai-era/) Justin Flitter · 7 September 2025 Read article → Gen AI ### [What is 'Authenticity' in the AI era?](https://www.newzealand.ai/what-is-authenticity-in-the-ai-era/) Justin Flitter · 2 September 2025 Read article → Insights ### [This made-with-AI Ikea commercial is brilliant.](https://www.newzealand.ai/this-made-with-ai-ikea-commercial-is-brilliant/) Justin Flitter · 22 July 2025 Read article → Best Practice ### [Using AI right now. How I'm doing it, and how you can too.](https://www.newzealand.ai/using-ai-right-now-how-i-m-doing-it-and-how-you-can/) Justin Flitter · 24 June 2025 Read article → Guidelines ### [Clarifying Trust Concerns With Popular AI Assistants](https://www.newzealand.ai/clarifying-trust-concerns-with-popular-ai-assistants/) Justin Flitter · 11 May 2025 Read article → Prompts ### [Dropping this AI Prompt Guide, for us to refer back to](https://www.newzealand.ai/dropping-this-ai-prompt-guide-for-us-to-refer-back-to/) Justin Flitter · 23 April 2025 Read article → Basics ### [What is this 'MCP' thing AI experts are talking about?](https://www.newzealand.ai/what-is-this-mcp-thing-ai-experts-are-talking-about/) Justin Flitter · 30 March 2025 Read article → Insights ### [ChatGPT, Perplexity, Claude, M365 Copilot - Which one provided the best response to this advanced prompt?](https://www.newzealand.ai/chatgpt-perplexity-claude-m365-copilot-which-one/) Justin Flitter · 28 March 2025 Read article → Gen AI ### [5 Reasons Plato Would Be Skeptical About AI](https://www.newzealand.ai/5-reasons-plato-would-be-skeptical-about-ai/) Justin Flitter · 26 March 2025 Read article → Basics ### [Try this Competitor Analysis Prompt](https://www.newzealand.ai/try-this-competitor-analysis-prompt/) Justin Flitter · 17 March 2025 Read article → The AI Academy ## Keep up with AI, *without keeping up with all of it.* Join the AI Academy for free and get our weekly newsletter on the most useful applications of AI for business. [Join the Academy →](https://academy.newzealand.ai/) --- # Our Partners | AI New Zealand Source: https://www.newzealand.ai/our-partners/ Our partners # The people we build *AI capability* with. We're proud to work with these organisations to advance your AI adoption. Aotearoa ## Local Partners. AI powered software and services made in New Zealand. - Arcanum - Autohive - Smartspace - Supahuman - Custom D - Element X - Ask Astro - Actvo - Law Cyborg - Tutaki - Hudson Gavin Martin - Custom365 Membership bodies ## Industry Partners. Industry organisations investing in governance and training for their members. - NZISM - EMA - Business Mentors - NZIOB - AQA - RMLA Platforms ## Global Technology Platforms. These are a handful of the organisations we're incredibly proud to have worked with. - OpenAI - Anthropic - Perplexity - AWS - Gemini - Microsoft ## Want to partner with us? If you are building AI capability for New Zealand businesses, we would like to hear from you. [Book an AI Chat →](https://www.newzealand.ai/contact-us/) [Join the Academy](https://academy.newzealand.ai/) --- # The Organisational Brain Toolkit | AI New Zealand Source: https://www.newzealand.ai/organisational-memory/ Free toolkit # Your team briefs AI from zero *every single time.* The Organisational Brain Toolkit is a folder structure, eight file templates, a governance folder and a capture method, so the context your business runs on is written down once and used by everyone. Not a database project. About an hour to start. Get the toolkit → See what is inside **8** file templates **5** governance templates **1** hour to a working set **65** minutes to capture an expert Claude and Copilot checklists The idea ## Every new chat starts from zero. The fix is not a data project and it is not a new tool. It is about eight plain documents, written once, pointed at the AI your team already has. The prompt lives for one message. The conversation dies with the tab. What survives is the knowledge layer underneath, and almost nobody is building it. That is the difference between AI being useful to whoever writes the best prompt, and AI being useful to your whole team. LAYER 1 ### The prompt. Lives one message. Gone the moment you send the next one. LAYER 2 ### The conversation. Dies with the tab. Nobody else in the business ever sees it. LAYER 3 ### Project knowledge. Every chat starts briefed. This is where the value is, and where the work goes. LAYER 4 ### Grounded org knowledge. Live sources, cited answers, permissions apply. What is inside ## Four kits, one download. Everything from the session, in the form you would actually use it. ### The Starter Memory Pack. Eight templates covering the minimum context an AI tool needs to be useful to the whole team. - Company profile - Tone and voice - Offers and pricing - Clients and ideal client profile - Standards and definition of done - Glossary and naming - Project instructions - Knowledge base rules ### The Governance Folder. Five templates for the rules the AI works inside, with sensible defaults already written. Every file in the pack takes its classification from here. - AI guidelines - Approved AI tools - Data classes - Agent permissions - AI incidents ### The Expert Capture Kit. The 65 minute method for getting a process out of somebody's head and into a document a new starter could follow. - The capture method, start to finish - The transcript to SOP prompts - The four questions to ask at each decision point ### Setup Checklists. Standing it up in the tools you already pay for, with the limits that actually bite. - Claude Projects - Copilot notebooks with SharePoint - File size and retrieval thresholds The method ## Seven rules, and your first hour. ### Seven rules for every file. 1. One topic per file. If the title needs an "and", split it. 2. One to three pages. Tight beats long. 3. Headings that describe content, because retrieval splits on heading structure. 4. Plain text, Markdown or clean Word. No decorative tables. 5. No scans, and no screenshots carrying meaning. An agent cannot read them. 6. A header block on every file: owner, last reviewed, review by, status, classification. 7. Consistent naming. Dates and versions live in the header, not the filename. ### Your first hour. | Time | File | | --- | --- | | 10 min | Company profile | | 10 min | Tone and voice | | 10 min | Offers and pricing | | 10 min | Clients and ideal client profile | | 5 min | Standards and definition of done | | 5 min | Glossary and naming | | 5 min | Project instructions | | 5 min | Knowledge base rules | Dictate for two minutes each and let AI structure it. You are editing, not authoring. Then the governance folder: five short files, about forty minutes, done before you connect anything that can send, publish or change a record. Before you start ## The five ways this fails. Boiling the ocean instead of writing eight files this week. Dumping the whole shared drive in, which buries the good material. No owner and no review date, so it quietly rots. Grounding an agent on a site nobody has run a permissions review over. Treating it as an IT project when the knowledge owners should write it. Get the toolkit ## Send it to me. Tell us where to send it and the download starts on the next page. The templates are yours to edit, rename and put to work. We will email you occasionally about practical AI for New Zealand businesses. Unsubscribe any time. Keep going ## The AI Toolkit webinars. This toolkit came out of one of them. Every week we take a working problem, build the thing live, and hand over what we used. Premium members get the full back catalogue and the resource library that goes with it. ### Premium membership. - Weekly AI Toolkit webinars, live and recorded - The full resource and template library - Every session's working files, like this one [See Premium membership →](https://www.newzealand.ai/premium/) --- # The Sales Call Coach Toolkit | AI New Zealand Source: https://www.newzealand.ai/sales-call-coach/ Free toolkit · The AI Toolkit # Your sales manager hears *one call in twenty*. Every call your team records is already transcribed. This toolkit turns those transcripts into coaching, scored against your sales method rather than generic advice, with the personal and commercial detail stripped before anyone reads it. Get the toolkit → See what is inside **3** platform tracks **2** standard templates **4** transcripts with planted faults **1** printable checklist The one idea ## The coach is only as good as your standard, and your standard needs anchors. Ask any AI tool to review a sales call and it will produce something fluent, confident and generic, because it is answering from the internet rather than from how you sell. The fix is not a cleverer prompt. It is a file that says what good looks like, specifically enough that two people reading the same transcript would agree. That is the difference between a rubric nobody can apply and one an AI can score line by line. | Line item | What most rubrics say | What an anchor says | | --- | --- | --- | | Rapport | Builds good rapport with the prospect. | Uses the prospect's name plus a business specific the rep looked up before the call, for example the herd size or the site they operate. | | Discovery | Asks effective questions. | At least four open questions, and at least one follow-up that goes a level deeper into an answer rather than moving to the next checklist item. | | Listening | Listens actively. | Plays the situation back in the prospect's own words and checks it, then refers to an earlier answer later in the call. | | Next step | Agrees next steps. | A date, a time and the names of who else will be in the room. "I'll send you an email" scores zero. | What is inside ## Three parts, one download. Everything you need to build it this week, and the governance to run it on real people without regretting it. ### Your standard. The two files that hold how your business sells, written the way an AI tool needs to receive it. About two hours with your sales manager. - What good looks like, with 0, 1 and 2 anchors - The scorecard every output must follow - The signals a rep must never miss - Your hard rules, and the score cap they trigger - What never reaches the report ### The platform tracks. Build it wherever your business already works. One track is enough, and the instructions are identical across all three. - Claude: a Project, then a shareable skill - Microsoft 365 Copilot: an agent on SharePoint - ChatGPT: a Project, and what is being retired - The instructions, version 1 and hardened version 2 - Five prompts, including the redaction audit ### A worked example. A complete fictional solar company, so you can test your build before you write a word of your own standard. - Four Teams transcripts with faults planted in them - The full standard, in markdown and .docx - Four real outputs, unedited - The answer key: what a good run finds - A printable two page checklist Three platforms ## Build it where your team already works. The method does not change between these. Your standard, the instructions and the tests are identical. Only the setup differs, and each track is written out step by step, checked against the vendor's own documentation in September 2026. | | Claude | Microsoft 365 Copilot | ChatGPT | | --- | --- | --- | --- | | Best when | You want it working this afternoon, and later shared as a skill. | Your transcripts, your standard and your people are already in Microsoft 365. | Your team is already on ChatGPT and you want one place for it. | | Set-up | A Project, eight knowledge files, paste the instructions. About 20 minutes. | An agent grounded on a SharePoint folder. One to two hours, plus indexing time. | A Project with project files and instructions. About 45 minutes. | | Watch for | Projects need a paid plan, and sharing needs Team or Enterprise. | It will not read markdown at all, so convert your standard to .docx first. | Custom GPTs are being retired, so build a Project rather than a GPT. | | Your data | Business plans do not train on your content. Check the plan you are on. | Stays in your tenant, under your existing labels and retention. | Training is off by default on Business and Enterprise, on by default elsewhere. | Before anyone trusts it ## Four tests, half an hour. Most AI workflows go into use because the first answer looked good. These are the checks that tell you whether it will still be right on the fiftieth call, and the pack includes transcripts with known faults so you can run them properly. TEST 01 ### Accuracy Write down what your sales manager catches. Then compare. Did it find every breach and every missed opportunity, or just the obvious ones? TEST 02 ### Drift Score the same call twice in two fresh chats. More than five points apart and you show bands, not numbers, until the anchors are tighter. TEST 03 ### Leak Search every output for anything on your exclusions list, including inside the quotes. This is where it usually goes wrong. TEST 04 ### Calibration Monthly, the manager scores three calls by hand and compares. It is how you notice the standard drifting before anyone else does. Before you start ## The five ways this fails. Going straight to the tool and never writing the standard, so the coaching is generic and the reps ignore it. Rolling it out without telling the team their calls are being assessed by AI, and what for. Showing a rep a score that a second run would not reproduce. Letting the number become the point, so someone ranks people in a spreadsheet instead of coaching them. Putting the prospect's health, finances or another customer's pricing into a document your team forwards around. Get the toolkit ## The Sales Call Coach Toolkit, free. Two standard templates, three platform tracks, the instructions and prompts, a complete worked example with four transcripts and an answer key, the governance templates, and a printable checklist. ZIP, about 780 KB. We will email you occasionally about practical AI for New Zealand businesses. Unsubscribe any time. Build the next one with us ## This is one workflow. The method works on any of them. The six judgement calls in this pack are not about sales. What is good? What is excluded? Is version one good enough? How much drift is acceptable? Who sees what? What would make you refuse? Run that sequence on any workflow your business repeats and you will build something your team actually trusts. We build these live with members every Friday, on real workflows, in 45 minutes. ### The AI Toolkit, weekly A 45 minute workshop every Friday, teaching practical AI skills for business. Members get the full resource library, every toolkit and the recordings. [AI Academy Premium →](https://www.newzealand.ai/premium/) Also free: the Organisational Brain Toolkit and the On-Brand Documents Toolkit, at newzealand.ai/toolkits. --- # The On-Brand Documents Toolkit | AI New Zealand Source: https://www.newzealand.ai/on-brand-documents/ Free toolkit # Your AI writes good documents. They just don't look like *your business.* The On-Brand Documents Toolkit is the setup that gets Microsoft Copilot and Claude applying your fonts, colours, logo and layout to everything they produce. Two standards files, three platform tracks, and the admin steps to turn it on for everyone rather than just you. Get the toolkit → See what is inside **22** files **3** platform tracks **2** page printable checklist Copilot, Claude Design and Claude skills The idea ## You are describing your brand in the prompt. That is why it does not work. Ask any AI tool for a proposal "in our brand colours, forest green and ochre gold, Georgia headings and Calibri body" and you get content that is fine and formatting that is generic. Typed brand instructions do not reliably become actual Word styling. They never have. What works is giving the tool something to read. Your brand standards as a file, and a document that already has the styles in it. Then every prompt inherits them and you stop retyping your own brand at your own software. LAYER 1 ### The prompt. "Use our brand colours." Lives one message, and Word mostly ignores it. LAYER 2 ### The brand kit. Configured once per platform. Real, but only reaches where that platform reaches. LAYER 3 ### The standards file. One file both platforms read. Survives whatever either vendor ships next. LAYER 4 ### The styled template. The file every document starts from. Does more work than the three above it. What is inside ## Three parts, one download. Everything from the session, in the form you would actually use it. ### The standards files. Two templates that hold your brand, written the way an AI tool needs to receive it. Hex codes and style names, not adjectives. - Brand and visual standards - Document standards and structures - A completed worked example - A styled Word template to pull apart - Numbered to continue the Organisational Brain Toolkit ### The platform tracks. Wiring each platform to read those files, including the admin steps most people never get to. - The Copilot official Brand Kit, end to end - Both Microsoft policies, and the trap between them - Claude Design, published organisation wide - A brand skill, with a starter SKILL.md - Where each one reaches, and where it does not ### The format explainers. What to do differently, and what to check, in each of the formats your business actually sends. - Word: proposals, SOWs and SOPs - PowerPoint: decks and the Slide Master rule - PDF, export and the traps in between - Baseline prompts, deliberately short - A printable two page checklist The method ## Seven rules, and your first afternoon. ### Seven rules that do the work. 1. Write the standards down. Hex codes and font names, never adjectives. 2. Never start from a blank document. Start from a file that already has the styles. 3. Save templates as .docx. Copilot in Word does not read .dotx files at all. 4. Link fonts and colours into the Slide Master, not onto individual slides. 5. Publish it to everyone. Until you do, you have solved this for one person. 6. Brand the source file, then export. There is no separate branded-PDF step. 7. Check the first three documents. Then trust the fourth. ### Your first afternoon. | Time | What you do | | --- | --- | | 30 min | Write the brand and visual standards | | 20 min | Write the document standards | | 40 min | Build one .docx with real named styles | | 30 min | Set up your platform track | | 20 min | Test it on a document you already sent | | 24 hrs | Wait, if you are on the Microsoft track | Copy the hex codes out of your brand guidelines rather than sampling them off a logo. A few points out shows up on every document from then on. Both platforms ## Which one, for what. Most businesses we work with end up running both. The pack covers both, and the standards file is what makes that work, because both platforms are reading the same thing. | | Microsoft Copilot | Claude | | --- | --- | --- | | Reach | Strongest in the Create tab and PowerPoint. Word, Excel and chat are partial. | Any output in any format, including web pages, with no per-format setup. | | Set-up cost | An admin, two policies, a paid Copilot Premium licence tier. | One toggle, then one person builds the design system or the skill. | | Best at | Sitting inside the apps your team already has open all day. | Portability. One brand setup covers docx, pptx, pdf and html. | | Watch for | It will not pick a brand kit from your prompt, and there is no version history. | Claude Design is still a beta. Large codebases and shared editing are rough. | Before you start ## The five ways this fails. Describing the brand in the prompt every time instead of writing it down once. Building a personal brand kit and stopping, which fixes this for exactly one person. Pointing Copilot at a .dotx template, which it silently ignores with no error. A deck template whose fonts were typed onto slides instead of set in the master. Choosing a licensed brand font and expecting AI to render it on everyone else's machine. Get the toolkit ## Send it to me. Tell us where to send it and the download starts on the next page. The templates are yours to edit, rename and put to work. We will email you occasionally about practical AI for New Zealand businesses. Unsubscribe any time. Keep going ## The AI Toolkit webinars. This toolkit came out of one of them. Every week we take a working problem, build the thing live, and hand over what we used. Premium members get the full back catalogue and the resource library that goes with it. The natural next one is the [Organisational Brain Toolkit](https://www.newzealand.ai/organisational-memory/). The two standards files in this pack are numbered to slot straight into it. ### Premium membership. - Weekly AI Toolkit webinars, live and recorded - The full resource and template library - Every session's working files, like this one [See Premium membership →](https://www.newzealand.ai/premium/) --- # The AI Champions Committee Resource Pack | AI New Zealand Source: https://www.newzealand.ai/ai-champions-committee/ Free resource pack # Everyone wants an AI committee. *Almost nobody writes down what it can decide.* The AI Champions Committee Resource Pack is a complete, ready-to-run programme for standing up AI governance and adoption inside a New Zealand organisation of 50 to 250 people. The mandate, the charter, the seats, the register, the tool framework, the policy and the training. You need a sponsor with authority, seven to nine people who will turn up, and ninety days. Get the resource pack → See what is inside **12** documents **115** page PDF **12** fill-in templates Every claim sourced The one decision that matters most ## You are building two things, not one. Conflating the committee with the network is the most common structural error, and it is the difference between a body that ships and a body that becomes the bottleneck. A 25 person group cannot make a timely decision. A network with approval rights becomes a shadow bureaucracy. | | The AI Champions Committee | The AI Champions Network | | --- | --- | --- | | Purpose | Decide | Diffuse | | Size | 7 to 9 seats | 2 to 3 percent of headcount | | Selected by | Role, not personality | Peer nomination | | Meets | Monthly, 90 minutes | Monthly, informal, 45 minutes | | Authority | Approves tools, sets policy, tiers risk, owns the register | None. Zero approval rights | | Time cost | 4 to 6 hours a month per member, 6 to 8 for the Chair | 2 to 4 hours a week, manager approved | | Fails when | It has no written decision rights, or everything routes through it | It has no visible executive backing | Citi's programme works precisely because its roughly 4,000 volunteer participants have no approval authority: the decisions sit elsewhere. A global financial services firm studied by MIT CISR routed everything through its AI review committees, took six months to approve low risk proposals, and watched shadow AI return until the board forced a redesign. [MIT CISR, Minimum Viable Governance for Generative AI, 19 March 2026](https://cisr.mit.edu/publication/2026_0301_GenAIGovernance_VanderMeulenJewerLevallet) What is inside ## Twelve documents, in the order you need them. Not a framework diagram. The artefacts you produce along the way, each one ready to fill in and take to your executive. ### Stand it up. Weeks one to three. The mandate, the charter and the seat map. 1. **00**Start here, and how to use the pack 2. **01**The Business Leader's Implementation Guide 3. **02**Terms of Reference and Charter Template 4. **03**Roles, Seats and the Champions Network ### Run it. Month two onward. The engine, and the rhythm that keeps it turning. 1. **05**AI Use Case Register and Prioritisation Playbook 2. **06**AI Tool Assessment and Approval Framework 3. **07**Drafting the AI Guidelines and Workplace AI Policy 4. **10**Operating Rhythm, Meeting Packs and Metrics ### Ground it. The induction reading, the capability build and the evidence. 1. **04**The Compliance and Standards Foundation 2. **08**Employee AI Training and Literacy Programme 3. **09**Selecting External Advisors and Service Providers 4. **11**Source Library and Verification Notes **And the parts you fill in come as their own files.** A PDF is no use for an artefact you have to complete, so all twelve of them are extracted as editable documents: the Terms of Reference, the champion role description, the use case intake form and register schema, the scoring method, the tool assessment checklist, the tool decision record, the monthly meeting pack and dashboard, the quarterly board report, the incident report form, the guidelines outline and the policy sections. Square brackets mark what you replace. The plan ## The first ninety days. One executive sponsor with real authority, a chair who is not the sponsor, and about half a day a week of coordination. Every stage ends in something you can put in front of your board. 1 Days 1 to 14 ### Mandate. A written statement from the CEO naming the committee, its authority and its sponsor. One paragraph, containing the words "is authorised to approve". Then the four questions the sponsor answers before anything else happens. **Out:** the mandate paragraph and a draft Terms of Reference 2 Days 15 to 30 ### Seats and baseline. Fill the seats by role, not by volunteer, because role based membership survives turnover. Then take a before picture: tool discovery, a five question anonymous pulse, and an honest governance inventory. **Out:** signed Terms of Reference, seat map, baseline report 3 Days 31 to 60 ### First decisions. Meeting one constitutes and tiers. Meeting two approves the first tools. A committee that approves nothing in its first two months loses credibility it will not recover. In parallel, draft the guidelines and nominate the network. **Out:** approved tools list, guidelines draft, champions named 4 Days 61 to 90 ### Operate. Publish the guidelines with a message from the sponsor, not from IT or Legal, and make it celebrate AI use rather than merely permit it. Run the first training. Triage the register down to three to five live use cases. Report to the board. **Out:** guidelines live, training delivered, first board report Why now, in New Zealand ## Adoption is high. Governance is not. New Zealand has deliberately chosen not to legislate specifically for AI. That does not mean there is no obligation. It means the obligations arrive through the Privacy Act 2020, the Fair Trading Act 1986, the Human Rights Act 1993 and employment law, and nobody is going to hand you a checklist. 13% of New Zealand companies had audit assurance or governance frameworks, while 66 percent were already using AI. Survey of 200 senior managers, August 2024, so treat it as directional rather than current. [Datacom State of AI Index →](https://datacom.com/nz/en/solutions/experience/insights/state-of-ai-index-ai-attitudes) 62% of New Zealanders would stop using a service if they were concerned about how that organisation was using AI. Nationally representative, 1,001 people, fieldwork March 2026. [One NZ AI Trust Report 2026 →](https://media.one.nz/ai-trust-report-2026) 47% of AI users were on personal, unmanaged accounts by October 2025, down from 78 percent a year earlier, as organisation managed adoption rose from 25 to 62 percent. Provision beats prohibition. [Netskope Cloud and Threat Report 2026 →](https://www.netskope.com/resources/cloud-and-threat-reports/cloud-and-threat-report-2026) 3.5 use cases pursued at once by leading organisations, against 6.1 for everyone else, while anticipating 2.1 times the return. Survey of more than 1,800 C-suite executives. Fewer bets win. [BCG AI Radar 2025 →](https://www.bcg.com/publications/2025/closing-the-ai-impact-gap) Before you start ## How this goes wrong, and how you will know. Every failure mode in the pack is documented, with the early warning signal that tells you it is happening. Six of the eleven are below. ### No written decision rights. Quorum, authority and escalation go into the Terms of Reference on day one. Signal: someone asks "can we just approve this?" and nobody knows ### Everything routes to the committee. Three tier routing, with asynchronous review of simple intake forms. Signal: low risk decisions creeping past two weeks ### Governance with no sanctioned tool. Banning tools does not reduce shadow AI. Providing better ones does. Signal: personal account use stays flat or rises ### The register dies at three months. A named owner on every entry, a mandatory look back, and publish it internally. Signal: no new entries, no closures, last update was month one ### No workflow redesign. Redesign has the single biggest effect on financial impact, and about one organisation in five does it. Signal: adoption high, hours saved high, nothing measurably changed ### The committee is captured. Role based seats, and a chair from the business rather than from IT. Signal: every decision reflects one function's priorities Get the pack ## Send it to me. Tell us where to send it and the download starts on the next page. One ZIP: the pack as a branded PDF to read and forward, the twelve documents as editable files, and the templates. Yours to use and circulate inside your organisation. We will email you occasionally about practical AI for New Zealand businesses. Unsubscribe any time. If you would rather not run it alone ## We can chair the first one. The pack is deliberately complete: a sponsor, seven to nine seats and ninety days is enough to run the whole thing without buying anything. Where we help is the part that is hard from the inside. Facilitating the first session so it produces decisions rather than a discussion, running the register at pace, and training the network so it diffuses rather than dissolves. Document 09 tells you how to evaluate us, and everyone else, on the same criteria. ### Where to next. - Book an AI Chat and talk through your seat map - Weekly AI Toolkit webinars, live and recorded - The full resource and template library [Book an AI Chat →](https://www.newzealand.ai/contact-us/) --- # Free AI Toolkits for New Zealand Businesses | AI New Zealand Source: https://www.newzealand.ai/toolkits/ Toolkits # The templates and methods, *free to take.* Every toolkit here came out of real work with New Zealand businesses, usually an AI Toolkit webinar. Practical files you can edit, rename and put to work this week, not another framework diagram. Available now ## Download a toolkit. Sales and coaching ### The Sales Call Coach Toolkit Turn recorded sales calls into coaching scored against your own method. Standard templates with anchors, three platform tracks, and a worked example with faults planted in it. - 2 standard templates, with scoring anchors - Claude, Copilot and ChatGPT tracks - 4 Teams transcripts and an answer key - Privacy Act, policy and register templates [Get the toolkit →](https://www.newzealand.ai/sales-call-coach/) ZIP, 780 KB Brand and documents ### The On-Brand Documents Toolkit Getting Copilot and Claude to apply your fonts, colours, logo and layout to everything they produce. Standards templates, both platforms, and the admin rollout. - 2 brand standards templates - Copilot Brand Kit, Claude Design and skills - Word, PowerPoint and PDF explainers [Get the toolkit →](https://www.newzealand.ai/on-brand-documents/) ZIP, 240 KB AI governance ### The AI Champions Committee Resource Pack The complete programme for standing up AI governance and adoption inside a 50 to 250 person business. Charter, seats, register, tool framework, policy, training. - A 115 page pack, plus editable files - 12 fill-in templates - A ninety day implementation plan [Get the pack →](https://www.newzealand.ai/ai-champions-committee/) ZIP, 1.2 MB Organisational memory ### The Organisational Brain Toolkit The folder structure, file templates, governance folder and capture method that stop your team briefing AI from zero every time. About an hour to a working set. - 8 file templates - 5 governance templates - The 65 minute expert capture method - Claude and Copilot setup checklists [Get the toolkit →](https://www.newzealand.ai/organisational-memory/) ZIP, 2.3 MB The AI Academy ## Keep up with AI, *without keeping up with all of it.* Join the AI Academy for free and get our weekly newsletter on the most useful applications of AI for business. [Join the Academy →](https://academy.newzealand.ai/) --- # Terms of Service | AI New Zealand Source: https://www.newzealand.ai/terms-of-service/ Legal # Terms of Service. **Effective** 1 December 2025 **Last updated** 8 August 2026 ## 1. Introduction and acceptance Welcome to [newzealand.ai](https://www.newzealand.ai/). These Terms of Service ("Terms") govern your access to and use of the website, community platform, workshops, courses and consulting services (collectively, the "Services") provided by **AI New Zealand Limited** ("we", "us" or "our"). By accessing or using our Services, creating an account, or purchasing a membership or ticket, you agree to be bound by these Terms and our [Privacy Policy](https://www.newzealand.ai/privacy-policy/). If you are using these Services on behalf of a business or organisation, you represent that you have the authority to bind that entity to these Terms. If you do not agree to these Terms, you must not access or use our Services. ## 2. Services and access AI New Zealand Limited provides advice, workshops and training to businesses in New Zealand. ### 2.1 Eligibility Our Services are intended for business users and professionals. By using our Services, you represent that you are at least 18 years old and have the legal capacity to enter into a binding contract under New Zealand law. ### 2.2 Account registration To access certain features, such as the community platform or course materials, you may be required to register an account. You agree to: - Provide accurate, current and complete information. - Maintain the security of your password and account details. - Notify us immediately of any unauthorised use of your account. ### 2.3 The AI Academy platform The AI Academy runs on **Circle**, a third party platform, at a separate address from this website and with its own account and login. Your use of the Academy is subject to Circle's own terms as well as these ones. We are responsible to you for the membership we sell and the content we publish there, but we do not control the platform itself and cannot promise it will be available without interruption. We may add, change or remove Academy features, spaces and content as the community develops. If we make a change that materially reduces what your membership gives you, you may cancel under clause 4.3. ## 3. Privacy and data protection Your privacy is paramount. Our collection, use and disclosure of your personal information are governed by our [Privacy Policy](https://www.newzealand.ai/privacy-policy/), which is incorporated into these Terms by reference. ### 3.1 Compliance with New Zealand law We comply with the **Privacy Act 2020**. By using our Services, you consent to the data practices described in our [Privacy Policy](https://www.newzealand.ai/privacy-policy/). ### 3.2 Data minimisation and rights We collect only the information necessary to provide our Services. You retain rights to access, correct or request deletion of your personal data as outlined in our [Privacy Policy](https://www.newzealand.ai/privacy-policy/). ### 3.3 Third party processing You acknowledge that certain data is processed by third party providers and is subject to their privacy policies in addition to ours: payment and subscription information by **Stripe**, and your Academy account, profile and posts by **Circle**. ## 4. Fees, payments and refunds ### 4.1 Pricing and payment Prices for workshops, courses and memberships are in New Zealand Dollars (NZD) unless otherwise stated. We use **Stripe** to process payments securely. We do not store your full credit card details on our servers. ### 4.2 Workshops and events - **Cancellations by you:** unless otherwise specified on the event booking page, cancellations made less than 7 days before an in-person workshop are non-refundable. - **Cancellations by us:** if we cancel an event, you are entitled to a full refund or a credit toward a future event. ### 4.3 Premium Membership Premium Membership gives you access to the AI Academy and the benefits listed on the [membership page](https://www.newzealand.ai/premium/). It is an individual membership, sold as either a monthly or an annual subscription, at the price shown on that page at the time you subscribe. - **Free trial:** membership starts with a 7 day free trial. Your card details are collected when you sign up but are not charged during the trial. At the end of the 7 days the subscription begins and the first payment is taken, unless you cancel before then. The trial is available once per person. - **Billing:** the subscription is billed in advance for the term you choose, monthly or annually, and **renews automatically at the end of each term until you cancel it**. A receipt is issued by email for every payment. - **Cancelling:** you can cancel at any time from your Academy account settings. Your access continues until the end of the term you have already paid for, and stops at that point. We do not refund the unused part of a term, so cancelling an annual subscription part way through leaves the rest of that year available to you rather than refunded. - **Failed payments:** if a payment cannot be collected we may retry it, and may suspend your access until it is paid. - **Price changes:** we will give you at least **30 days notice by email** before a change to the membership price applies to you, so you can cancel before it takes effect if you would rather not continue at the new price. - **One person per membership:** Premium Membership is for you individually. You may not share your account or your login with anyone else. Recorded webinars and workshops in the Academy are provided to be watched in the Academy. They are not available to download, and clause 5.1 applies to them. ## 5. Intellectual property ### 5.1 Our content All content provided by AI New Zealand Limited, including but not limited to course materials, slide decks, frameworks, logos and website copy, is the exclusive intellectual property of AI New Zealand Limited or its licensors. You are granted a limited, non-exclusive, non-transferable licence to use these materials for your own personal or internal business learning purposes. **You may not resell, redistribute or reproduce our materials without our express written consent.** ### 5.2 Third party presenters Workshops or courses may feature content presented by third party experts. Intellectual property rights for such content remain with the respective presenters, and you are granted a licence to use such content solely for the purpose of the specific workshop or course. ### 5.3 User generated content By posting content, including posts, comments and images, to our community spaces: - You retain ownership of your content. - You grant AI New Zealand Limited a non-exclusive, royalty-free, worldwide licence to host, display and reproduce your content within the context of providing the Service. - You represent that you have the right to post such content and that it does not infringe on the rights of others. ## 6. User conduct We are a professional community. You agree not to: - Harass, abuse or threaten other members or staff. - Post content that is unlawful, defamatory or promotes hate speech. - Use the Services to send unsolicited commercial messages (spam). - Scrape or collect data from our Services using automated means. - Attempt to bypass any security measures of the platform. - Share member-only content, recordings or materials outside the community, or give anyone else access to your account. - Record or republish live sessions without our written consent. - Use another member's personal information, or the member directory, to send unsolicited approaches. We reserve the right to suspend or terminate the accounts of users who violate these guidelines, with data retention or deletion handled in accordance with our [Privacy Policy](https://www.newzealand.ai/privacy-policy/). If we suspend or end a paid membership because you have breached these Terms, the remaining part of that billing period is not refunded. If we end your membership for any other reason, we refund the unused part of the period you have paid for. ## 7. Disclaimers and limitation of liability ### 7.1 General advice warning **The information provided through our workshops, courses, consulting and community is for educational and general informational purposes only.** It does not constitute legal, financial or specific organisational advice. - AI technology evolves rapidly; information accurate at the time of publication may become outdated. - You should obtain specific professional advice before making business decisions based on AI outputs or our content. ### 7.2 AI outputs You acknowledge that artificial intelligence tools discussed or demonstrated may produce inaccurate or "hallucinated" results. AI New Zealand Limited is not responsible for the output of any third party AI tools you use based on our training. ### 7.3 Consumer Guarantees Act Nothing in these Terms is intended to limit your rights under the **Consumer Guarantees Act 1993** if you are not acquiring the services for business purposes. However, if you are acquiring our Services for the purposes of a business, you agree that the Consumer Guarantees Act 1993 does not apply. ### 7.4 Limitation of liability To the fullest extent permitted by New Zealand law, AI New Zealand Limited shall not be liable for any indirect, incidental, special or consequential damages arising out of or in connection with your use of the Services. Our total liability to you for any claim arising out of these Terms shall not exceed the amount you paid to us for the Service in the 12 months preceding the claim. ## 8. Modifications to Terms We may update these Terms from time to time to reflect changes in our Services or legal requirements. We will notify users of material changes by email or a prominent notice on our platform. Continued use of the Services after such changes constitutes acceptance of the new Terms. ## 9. Governing law and dispute resolution These Terms are governed by and construed in accordance with the laws of **New Zealand**. - **Disputes:** we encourage you to contact us first to resolve any issues. - **Jurisdiction:** any legal disputes arising from these Terms are subject to the exclusive jurisdiction of the courts of New Zealand. ## 10. Contact us For any questions regarding these Terms or our Services, please contact us: **AI New Zealand Limited** Email: [justin@newzealand.ai](mailto:justin@newzealand.ai) [Contact us](https://www.newzealand.ai/contact-us/) --- # Privacy Policy | AI New Zealand Source: https://www.newzealand.ai/privacy-policy/ Legal # Privacy Policy. **Effective** 1 December 2025 **Last updated** 21 September 2026 ## 1. Who we are **AI New Zealand Limited** provides advice, workshops and training to businesses in New Zealand. This policy explains how we collect, use, store and disclose personal information, and what rights you have over it. We handle personal information in accordance with the **Privacy Act 2020** and its information privacy principles. This policy forms part of our [Terms of Service](https://www.newzealand.ai/terms-of-service/). ## 2. The short version We collect very little. We use Google Analytics to understand which pages are read and how people find us, which sets cookies. We run no advertising networks and no advertising pixels, and we do not sell personal information. The information you can give us directly through this site is what you type into the toolkit download form, the booking form on our contact page, and the review form we send to people reviewing a draft course. Everything else, joining the AI Academy or paying for a course, happens on a named third party platform with its own privacy policy. The **AI Academy** is the exception worth reading properly. It is a membership community on a separate platform, with its own account, its own subscription billing and content other members can see. Sections 3.3 and 3.4 cover it. ## 3. What we collect, and why ### 3.1 The toolkit download form This is one of three forms on this website; the others are the booking form in section 3.2 and the course review form in section 3.7. When you request a toolkit we collect: - Your name and work email address, so we can identify who requested it and send you related material. - Your LinkedIn profile URL, which is optional, to help us understand who our resources are reaching. - The page you submitted the form from and your browser's user agent string, so we know which toolkit was requested and can diagnose problems. We use this to send you the toolkit and occasional email about practical AI for New Zealand businesses. You can unsubscribe from that at any time, and unsubscribing does not affect anything you have already downloaded. ### 3.2 Booking a call When you book an AI Chat, Discovery Call or Planning Session on our contact page we collect your name, work email address, company, phone number if you give it, and anything you type in the agenda box. That record is stored by AI New Zealand in our database (Supabase, see section 6), the appointment is added to our Microsoft 365 calendar, and the meeting runs on Microsoft Teams. We use the details to run the meeting, send the invitation and a reminder, and to follow up afterwards. The confirmation email carries a link you can use to reschedule or cancel. We do not operate a general contact form on this site. ### 3.3 The AI Academy The AI Academy is a membership community running on **Circle**, a third party platform. It is a separate system from this website, with its own account and its own login. If you join, Circle holds the following for us: - **Your account:** name, email address and password, and the record of when you joined and last signed in. - **Your profile:** whatever you choose to add, such as a photo, headline, bio, location and links. - **What you post:** posts, comments, replies, reactions, direct messages to other members, event RSVPs, and your progress through any course you take. - **How you use it:** sign in records, the spaces and posts you open, and your notification settings. **The Academy is a community, so some of this is visible to other people.** Your profile, your posts and your comments can be seen by the other members of any space you belong to, and by us as administrators. Direct messages are visible to the member you send them to. Please treat anything you post there as something other members will read, and do not post confidential information about your organisation or anyone else's. Circle also sends email about activity in the community, such as replies to your posts and event reminders. Which of those you receive is yours to set, in your Academy notification settings. Circle processes this information under its own [privacy policy](https://circle.so/privacy) as well as under this one. ### 3.4 Membership subscriptions and payments Premium Membership is a paid monthly subscription. You subscribe through the Academy's checkout, which runs on Circle, and the payment itself is handled by **Stripe**. **Your card details go straight to Stripe.** They are typed into Stripe's own secure fields, they never pass through this website, and we never see or store your full card number. Because the membership recurs each month, Stripe keeps your payment method on file so it can take the next payment, and it stops doing so when you cancel. What we can see and keep is the record of the subscription itself: - Your name and email address, and the billing details you entered. - Which plan you are on, whether it is in trial, active, past due or cancelled, and the dates it started, renews or ended. - Your payment history and the receipts issued, including the amount, the date, and the card brand and last four digits. **Never the full card number.** Membership starts with a 7 day free trial. Your card is collected when you sign up but is not charged during the trial, and the first payment is taken at the end of it unless you cancel before then. Receipts are issued automatically by email. Stripe processes payment information under its own [privacy policy](https://stripe.com/privacy). It also uses that information to detect and prevent fraud, which is a purpose of its own as well as ours. ### 3.5 Email and correspondence If you email us, we keep the correspondence so we can respond and maintain a record of what was agreed. ### 3.6 Server logs and analytics Our host, **Netlify**, records standard technical information such as IP address, request time and page requested. This is used for security and to keep the site running, not to build a profile of you. We also use Google Analytics, which is described in section 4. ### 3.7 Reviewing a draft course We sometimes ask people to review a course before we build it, using a review page we send them a link to. That page collects your name, email address, organisation and role, whether you are happy to be named as a reviewer, and your answers: your rating of and comments on each section, what you think is missing, and what you would expect to pay. Your answers are saved to our database (Supabase, see section 6) as you type, and a copy is kept in your own browser so you can close the page and finish later on the same device. We use them to improve the course and to follow up with you about your feedback. We name you as a reviewer only if you tick the box that says we can. ## 4. Cookies and analytics We use **Google Analytics 4** to understand how the site is used: which pages are read, how long people stay, and how they arrived. This is the only tracking on the site. To do that, Google Analytics sets cookies in your browser that distinguish one visitor from another and group your page views into a session. It also records your approximate location, which Google derives from your IP address and does not store, along with your device and browser type and the page that referred you. We use this in aggregate, to decide what to write and what to fix. We do not use it to build a profile of you as an individual, we do not combine it with the details you give us through the toolkit form, and we do not use it for advertising. ### 4.1 How to opt out - Install Google's [Analytics opt-out browser add-on](https://tools.google.com/dlpage/gaoptout), which blocks it on every site. - Block cookies or use your browser's tracking protection. Blocking JavaScript costs you very little: every page still reads the same. Only the booking calendar and the course review form need it, and both pages give you an email address to use instead. **We run no advertising networks and no advertising or social media pixels**, and we do not sell or rent personal information to anyone. Third party platforms we link out to, including Circle, Microsoft and Stripe, set their own cookies once you are on their sites. Their policies apply there. ## 5. Who we share information with We share personal information only with the service providers that make our services work, and only as far as they need it: | Provider | What for | Where it is held | | --- | --- | --- | | **Supabase** | Our database, including toolkit form submissions, booking records, course reviews and client records | Singapore | | **Netlify** | Website hosting and server logs | United States | | **Circle** | The AI Academy membership community, and the checkout you subscribe through | United States | | **Stripe** | Payment processing, recurring subscription billing and fraud prevention | United States and global | | **Microsoft** | Email, calendar and Teams meetings | Australia and global | | **Google** | Website analytics | United States and global | We may also disclose information where the law requires it. We do not sell personal information. ## 6. Where your information is stored As the table above shows, some of our providers hold information outside New Zealand. Under the Privacy Act 2020, sending personal information overseas is treated as a disclosure, and we rely on those providers being subject to comparable privacy safeguards. If you would rather your information were not held offshore, contact us before using the relevant service, because for most of these there is no local alternative we can offer. ## 7. How long we keep it - **Toolkit and marketing contacts:** until you unsubscribe or ask us to delete your details. - **Client and engagement records:** for seven years after the engagement ends, which is the period our tax and business records obligations require. - **Academy accounts:** for as long as your account is open, then in line with Circle's deletion process. Posts and comments you made in the community are not always removed along with the account, so tell us if you want those taken down too. - **Subscription and payment records:** for seven years, for the same tax and business records obligations. This is the record of what was paid and when, not your card details, which we never hold. - **Server logs:** for the short period our host retains them. We do not keep personal information for longer than we need it for the purpose we collected it. ## 8. Keeping it secure Access to our systems is restricted to the people who need it. Our database enforces row level security, so the tables holding personal information cannot be read by the public website or by anyone without explicit authorisation. Data is encrypted in transit. No system is perfectly secure, and we cannot guarantee absolute security, but if a privacy breach occurs that is likely to cause serious harm we will notify you and the Office of the Privacy Commissioner as the Privacy Act requires. ## 9. Your rights Under the Privacy Act 2020 you have the right to: - **Ask what we hold** about you, and get a copy of it. - **Ask us to correct it** if it is wrong, or to attach a statement of correction if we disagree. - **Ask us to delete it**, which we will do unless we are required to keep it. - **Unsubscribe** from our email at any time. Email [justin@newzealand.ai](mailto:justin@newzealand.ai) and we will respond within 20 working days, which is the timeframe the Act sets. We do not charge for this. ## 10. Complaints If you think we have mishandled your personal information, please tell us first so we can put it right. If you are not satisfied with our response, you can complain to the Office of the Privacy Commissioner: [privacy.org.nz](https://www.privacy.org.nz) · 0800 803 909 ## 11. Children Our services are intended for business users and professionals, and are not directed at children. We do not knowingly collect personal information from anyone under 18. If you believe we have, contact us and we will delete it. ## 12. Changes to this policy We may update this policy to reflect changes in our services or the law. The effective date at the top of this page shows when the current version took effect. If we make a material change we will say so by email or a prominent notice on the site. ## 13. Contact us For any question about this policy or the information we hold: **AI New Zealand Limited** Email: [justin@newzealand.ai](mailto:justin@newzealand.ai) [Contact us](https://www.newzealand.ai/contact-us/) · [Terms of Service](https://www.newzealand.ai/terms-of-service/) --- # Aspire Property Management: an AI case story Source: https://www.newzealand.ai/aspire-property-management-case-story/ Client case story # A virtual property manager, built into Aspire's own portal. How Aspire Property Management gave its property managers an AI assistant grounded in their own SOPs, the Residential Tenancies Act and six insurance policies, across email, WhatsApp and maintenance. **Justin Flitter** · Founder, AI New Zealand · October 2026 · 6 min read Aspire Property Management looks after more than 1,100 residential properties across three cities, with a team of around fifteen property managers. Its growth was never limited by demand. It was limited by what each property manager could physically get through in a day. Over twelve weeks, from July to September 2026, we built an AI layer directly into the management portal Aspire already runs. Its email assistant, Daryl, has auto-drafted more than 8,000 replies since August, for around $10 a day, and every one of them still goes out under a property manager's name, after a property manager's click. ## The story at a glance - **~3,749 hours a year** of property manager time released across four workflows (scoped value, client verified). - **$299,920 a year** indicative value of that time, at $80 an hour (scoped value, client verified). - **120 to 180 properties** per property manager, the capacity target, with no added headcount (client verified). - **8,000+ emails** auto-drafted since August, reported from the production system on 17 September 2026 (client verified). ## The organisation Aspire is known for a high standard of owner and tenant communication, and for investing in the systems that make consistent service possible at scale. Unusually for a business its size, Aspire built and maintains its own management portal: a Supabase and React platform that pulls Tapi (maintenance), Renti (tenancy and financials) and per-property Google Drive folders into one working environment. That commitment to owning its technology is what made a genuinely embedded AI solution possible. ## The challenge At around 120 properties per person, much of every working day went on repetitive communication: replying to tenants and owners, chasing contractors, coordinating maintenance, and looking up the right clause in the Residential Tenancies Act or an insurance policy. Institutional knowledge sat in people's heads rather than in systems. There were no standardised email responses and no structured SOP library, so when a property manager left, the patterns and precedents they carried left with them. The path to growth was clear. The operating model made it hard to walk without simply adding headcount. ## The turning point Aspire's leadership recognised the bottleneck was structural, not a staffing problem. Lifting each portfolio from 120 towards 180 properties meant absorbing the repetitive first-touch work, not hiring more people to do it. The data was already there, in rich tenancy, maintenance and contact records inside Aspire's own portal. What was missing was an AI layer built directly into that system. One codebase, one system of record. ## The approach Discovery produced the roadmap in April 2026. The implementation was one combined twelve-week build in four phases, led by our consultant James Oldham with Aspire's Managing Director, Mike Atkinson, as sponsor. ## 1 A cleaner base to build on July. Source material handed over, and a structured review and tidy-up of the existing portal codebase. Starting here reduced risk for every phase that followed. ## 2 Meet Daryl July to August. A knowledge base of Aspire's SOPs, the RTA and six insurance policies, plus an auto-generated brand-voice guide. On top sits Daryl, the email assistant. It filters inbound mail, identifies the sender, retrieves the right source and drafts a reply in Aspire's voice. ## 3 The channel tenants actually use August. The same grounding, approval and audit logic extended to WhatsApp, with a verified WhatsApp Business profile. Maintenance requests and applications are routed straight to the right Tapi job or Renti listing. ## 4 Built to run without us September. A full-screen portal inbox, a shared WhatsApp inbox with per-property-manager filters, send-side safety gates, and a weekly optimisation review handed to Aspire on 8 September. ## Human in the loop, by design One rule was non-negotiable: a person approves every client-facing message. Daryl drafts, the property manager decides and sends. Edits are captured as training signal rather than thrown away. Before launch we also set a precision bar of 0.95 that any reply type must clear before it is ever allowed to send unattended. Trust is earned on evidence, not assumed. ## What changes for a property manager - **Email triage and drafting.** Before, every email was read in Gmail, history looked up by hand and replies drafted from scratch. Now mail is filtered and classified, the sender matched against Aspire's records, and a draft grounded in the right SOP, RTA clause or policy is waiting before the property manager opens it. - **One knowledge layer.** Before, the Act, the policies and per-property documents were not searchable, and tone and accuracy varied from person to person. Now one knowledge base, documented so Aspire can update its own SOPs and policies, powers every channel in one consistent voice. - **WhatsApp.** Before, tenant and tradie messages were handled ad hoc, often outside any system. Now a shared portal inbox identifies the sender and routes the message to the right property manager or an admin queue, with drafts from the same knowledge base as email. - **Maintenance and applications.** Before, a request meant manual look-ups and outreach across Tapi, Renti and Drive. Now replies carry a direct link into the relevant Tapi job or Renti listing, so the work starts in the platform that already runs it. ## Three judgement calls that shaped the build - **One system of record.** Built inside Aspire's portal rather than wrapping a platform around it. One codebase to maintain, nothing to keep in sync. - **Link, don't rewrite.** Tapi and Renti already do their jobs well. Linking into them keeps each platform's audit trail intact. - **Draft, then decide.** Automation targets the slow part, retrieval and drafting. The decision to send stays with the person who owns the relationship. ## The impact Across four workflows the scoped value is nearly 3,750 hours and $300,000 a year back in the business. Illustratively, that is about two full-time roles of capacity, or room for the same fifteen property managers to carry around 900 more properties at 180 each. - **Maintenance triage and contractor orchestration:** 45 to 10 minutes a task, ~1,517 hours and $121,360 a year. - **Email responder and triage (Daryl):** 15 to 5 minutes, ~1,300 hours and $104,000 a year. - **RTA, SOP and brand voice knowledge layer:** 30 to 5 minutes, ~542 hours and $43,360 a year. - **WhatsApp tenant and owner assistant:** 10 to 1 minute, ~390 hours and $31,200 a year. The biggest single win is maintenance. Every job used to mean chasing photos, checking obligations under the RTA and hunting for the right contractor. Cutting that from around 45 minutes to 10 is where more than $120,000 a year comes back. And this is not a pilot running on a handful of test emails. Daryl is filtering tens of thousands of real inbound emails and has drafted more than 8,000 replies since August, grounded in Aspire's own SOPs, the RTA and its insurance policies. ## How these figures were worked out Hours and dollar values are the per-workflow benchmarks from Aspire's roadmap, carried into the implementation proposal (revised 29 May 2026), at an $80 hourly rate. Time per task is from the AI New Zealand use case register. The FTE and property figures are illustrative, derived from those benchmarks. Draft volume and running cost are reported from the production system on 17 September 2026. Aspire's Managing Director, Mike Atkinson, reviewed this story and approved its figures in October 2026. Measured results follow at the 60-day review after go-live, and we will update this story when they land. Draft with AI. Decide as a person. Grow the portfolio, not the payroll. Curious what an assistant built into your own systems could do for your team? [Talk to us](https://www.newzealand.ai/contact-us/). ## Power up your potential with practical AI skills. Contact us to discuss how Artificial Intelligence could boost your business. [Book an AI Chat](https://www.newzealand.ai/contact-us/?service=ai-chat#book) [Explore the AI Operating System](https://www.newzealand.ai/ai-os/) --- # The AI integration layer nobody buys, and why it pays Source: https://www.newzealand.ai/integration-layer/ 3 August 2026 # Integration is the layer nobody buys. It's the only one that pays. Most AI programmes in New Zealand right now are one training session and a lot of hope. **Justin Flitter** · Founder, AI New Zealand · 3 August 2026 · 5 min read The session goes well. People leave energised. A few of them go home and do something genuinely impressive with it. Then six weeks later you ask how it is going and the answer is a slightly awkward "yeah, good, when I remember". That is not a motivation problem, and it is definitely not a people problem. It is a design problem. The AI is sitting in a different place from the work. ## The tab tax Think about what actually happens when AI lives in its own browser tab. Someone is in the job management system, or the practice software, or the ledger, doing a task they do every week. To use AI they have to stop, switch, copy the relevant context out, paste it in, prompt, read, and paste the result back where it belongs. Every step is small. None of them is hard. Together they cost a little act of willpower, and they cost it every single time. Willpower is not a strategy. Some weeks your team has it. Most weeks they have a deadline, a sick kid, and a client chasing a variation. So usage does not spread. It concentrates in the two or three people who were always going to push through the friction. Their output gets sharper. Everyone else drifts back to how they did it before. The gap inside your own business widens, quietly, month after month. That's not an adoption problem. It's an operating model problem. ## The four stages, and where most firms are stuck There is a fairly reliable arc that businesses move through. - **Individual experimentation.** A few people are using AI on the side of their actual job. Results vary wildly and nothing is shared. - **Policies and basic training.** A workshop runs. A policy lands. Behaviour barely shifts. - **Scattered tool adoption.** Copilot is switched on, someone has ChatGPT, someone else prefers Claude. This stage feels like progress and is actually the riskiest one. It is hard to govern and harder to measure. - **A shared** [**AI operating system**](https://www.newzealand.ai/ai-os/)**.** Your methodology, your voice and your workflows encoded once, connected to your real systems, used by everyone, improved every month. Almost every business I talk to is somewhere in stages two and three. And almost every one of them is planning more of stage two, because training is the thing that is easy to buy. Stage four is where the value compounds. The bridge between them is integration. ## What integration actually means An AI operating system has four layers. - **Skills** are the workflows configured for the outputs your team produces every week. Proposals. Client updates. Reports. Meeting minutes. - **Knowledge** is your methodology, templates, past work and tone of voice, captured once so quality does not depend on who happens to be at the keyboard. - **Integration** is the connection to the systems where work actually lives. Microsoft 365, SharePoint, your CRM, Teams, your industry platform. - **Governance** is the permissions, the versioning and the guardrails. AI drafts. Humans approve. Firms buy Skills readily because it feels like progress and you can see it in a calendar invite. Some buy Knowledge. Integration gets deferred, because it sounds like an IT project with a business case attached. It usually is not. Integration rarely means a rebuild. It means one workflow, one connection, one point where the AI reaches the system of record instead of a person copying and pasting into it. ## A scene you will recognise A mid-sized services business, around twenty staff, upper North Island. Good people, busy, well run. They had done the training. Two of the team were flying. The rest had gone quiet. The weekly bottleneck was not glamorous. It was turning site notes and email threads into client-ready updates. Roughly a day and a half of someone's week, spread across five people in fifteen minute chunks nobody was counting. The fix was not a better prompt. It was connecting the drafting workflow to the place the source material already lived, so the AI could see the notes and the past updates without anyone assembling a context pack first. The task did not disappear. It became review instead of production. AI drafts, a human applies judgement, the update goes out sharper than it used to and hours earlier. Not miraculous. Real. ## The economics, plainly Indicatively, we see somewhere in the order of $15,000 of recovered value per person per year when a team moves from scattered tools to a connected system. Treat that as a planning number, not a promise. Your mix of work will move it either direction. The important part is not the figure. It is where the figure comes from. It does not come from people being better at prompting. It comes from removing the friction between the thinking and the system, so the good work happens by default rather than by effort. The sweet spot is roughly 25 to 50 staff. Big enough to feel the cost of inconsistency, small enough to change how things run without a committee. ## What to do next week None of this needs a budget round to start. - **Count the copy and pastes.** Sit with one person doing one recurring task with AI. Every copy and paste is a place the work is leaking. - **Pick the single workflow with the most handoffs**, not the one with the most enthusiasm behind it. - **Ask what the AI would need to see** to do the first draft properly. That answer is your integration spec, and it is usually shorter than people expect. - **Name the number.** Cycle time, cost to serve, hours to client-ready. One lever, with a baseline. Move a metric, not a use case. - **Set the guardrail before you connect anything.** What the AI may draft, what a human must approve, what it must never touch on its own. ## The shift to make The next competitive edge for New Zealand businesses will not come from one person being brilliant at prompting. It will come from firms that have connected AI to how they actually work, so the capability belongs to the business rather than to whoever happened to attend the workshop. Skills get you started. Integration is where value compounds. That's an [operating system](https://www.newzealand.ai/ai-os/). That's the move. ## Power up your potential with practical AI skills. Contact us to discuss how Artificial Intelligence could boost your business. [Book an AI Chat](https://www.newzealand.ai/contact-us/?service=ai-chat#book) [Explore the AI Operating System](https://www.newzealand.ai/ai-os/) --- # AI is becoming operational infrastructure | AI New Zealand Source: https://www.newzealand.ai/operational-infrastructure/ Field Notes · The Great New Zealand AI Roadshow # Everyone has AI in the loop. The difference is the *operating model*. AI is no longer just about content generation. It is becoming operational infrastructure. **Justin Flitter** · Founder, AI New Zealand · 26 June 2026 · 3 min read For two years, "using AI" mostly meant generating content. A faster email. A tidier slide. A blog post in your own voice. Useful, but shallow. Across the Great New Zealand AI Roadshow, a different story kept showing up on stage. AI is moving out of the chat window and into the plumbing of the business. Justin Flitter, founder of AI New Zealand, named the shift. "A year ago, this was 'ask and act'," he told the room. "Now we can delegate the whole process to agents and skills: to research, to create, and to generate the output." The punchline landed hard: "Just having AI in the loop is not what sets you apart." > Just having AI in the loop is not what sets you apart. Justin FlitterFounder, AI New Zealand That reframes the work. Richard Kenyon, Associate Director of AI and Automation at Datacom, drew the line most businesses miss. "Adoption is using AI to automate the way we work today," he said. "It doesn't really make sense to automate the way we work today, because that's based on legacy tech. It's based on legacy processes." In his telling, infrastructure is no longer fixed. "A modern IT infrastructure is one that you are continuously changing, because we can." Once agents run real work, they need real engineering. Kenyon calls agent assurance "a new budget line item": sampling, audit and version control for software that drifts like a factory line. And infrastructure that acts has to be governed like infrastructure, not coached like a chatbot. Raymond Schippers from Check Point was blunt. The instructions you send an agent "are just suggestions and guardrails, like asking your teenage children to clean a room. It might work. It might not." Real control is technical: least privilege, human review, rollback. The stakes are operational, not cosmetic. Anchali Anandanayagam, Partner at Hudson Gavin Martin, pointed to Air Canada, held liable when its chatbot invented a policy that never existed. Her line for every board: "You can't outsource your accountability to these tools." None of this is really a technology problem. Asa Cox, founder of Numa, has run AI adoption work with more than 100 Kiwi businesses. His verdict is the one every leader needs to hear: "The technology is ready, the people are not." > The technology is ready, the people are not. Asa CoxFounder, Numa This is the move. Not better screenshots. An operating model where AI senses, thinks, decides and acts inside your systems, with your context encoded once and used by everyone. The differentiator was never the model. It is the *operating model*. The tools are ready. The infrastructure question is whether you are. ## Build the operating model, not just the screenshots. See where the Roadshow conversation goes next, or start mapping your own use cases and roadmap with the team at AI New Zealand. [Explore the Roadshow →](https://www.newzealand.ai/roadshow/) [Book a Discovery Call](https://www.newzealand.ai/contact-us/?service=discovery-call#book) Continue reading ## More from AI New Zealand. Client case story ### [A virtual property manager, built into Aspire's own portal.](https://www.newzealand.ai/aspire-property-management-case-story/) Read article → AI Operating System ### [Everyone is buying the skills layer. Almost nobody is buying the integration layer.](https://www.newzealand.ai/integration-layer/) Read article → Insights ### [What comes after LLMs now has a New Zealand face on it.](https://www.newzealand.ai/what-comes-after-llms-now-has-a-new-zealand-face-on-it/) Read article → --- # AI Gave Us the Answer. Critical Thinking Saved the Deal. Source: https://www.newzealand.ai/ai-answer/ Jan, 2026 · Leadership # AI Gave Us the Answer. Critical Thinking Saved the Deal. Critical thinking is your competitive moat. Not as a nice idea. **Justin Flitter** · Founder, AI New Zealand · Jan, 2026 · 5 min read I'm the CMO of a software development business in New Zealand. We build custom platforms for mid-market organisations - the kind that have real complexity, real customers, and real consequences when you get it wrong. And lately, like every other leadership team, we've had a new colleague in the room. AI. It shows up with confidence. It writes faster than any human. It turns messy notes into neat decks and neat decks into neat plans. It's unbelievably helpful. It's also often wrong in the only way that matters. Not factually wrong. Contextually wrong. ## The pitch we "almost" won. A few months back, we were in late-stage conversations with a New Zealand services business looking to rebuild a customer portal and integrate it with their back-office systems. Big project. High visibility. The kind of work you want on your website, your case studies, and your quarterly report. We'd done the workshops. We'd mapped the stakeholders. We'd even made friends with the CFO (always a good sign). Then the client went quiet. Not dead quiet. More like "we're suddenly very busy, thanks for the doc" quiet. So we did what modern teams do. We asked AI to help. We fed it the meeting notes, the discovery outputs, and our proposed scope. We asked it to produce a follow-up strategy: objection handling, likely concerns, messaging angles, and a re-engagement email sequence. The output was... good. Really good. It recommended we position around speed to value, reduced operational effort, and a cleaner customer experience. It suggested a crisp email with a soft close and three optional next steps. It even created a short one-page "why us" summary that looked like something a senior marketer had written on a great day. It felt like an answer. And that's the trap. AI commoditises answers. Everyone can get a decent follow-up email now. Everyone can get a plausible strategy slide. Everyone can sound smart on the internet. The real question is: **Is it the right answer for this situation, with these people, at this time?** That's not an AI problem. That's a judgment problem. ## The moment our Head of Delivery got uncomfortable. Before we hit send, I did what I've trained myself to do in the last year: I pulled someone from a different part of the business into the loop. Not to "approve the copy". To interrogate the thinking. Our Head of Delivery skimmed the AI-generated plan and said, "This is tidy... but it assumes the client's problem is speed and effort." He paused. "I don't think that's their real problem." Now, he wasn't guessing. He'd spent six years implementing systems for organisations like this. He'd lived through the politics, the workarounds, the late-night escalations, the 'we didn't realise finance did it that way' surprises. Deep domain knowledge. Lived experience. The stuff you don't get in a transcript. He pulled up one line from our notes that the AI had treated as background noise: "We've been burned before with a vendor who promised integration was simple." The AI had filed that under generic "risk objection". My colleague read it as trauma. Different interpretation. Different reality. So we rewrote our entire approach. Not the email. The strategy. ## Asking better questions (instead of sending better emails) We got on the phone with the client sponsor and didn't lead with our shiny solution. We led with a question: Last time you got burned, what actually happened?" It turned out the previous vendor hadn't just missed deadlines. They'd built something that looked right in demos, then fell apart under real usage. Data mismatches. Manual reconciliations. Frontline staff inventing spreadsheets to cope. Executives being promised one thing and getting another. The sponsor said, "It was embarrassing." That one word told us everything. This wasn't a sales cycle about features. It was a decision about professional risk. And suddenly, the AI's suggested message about "speed to value" wasn't just unhelpful. It was dangerous. It signaled we didn't understand the stakes. So we shifted. We positioned our team around integration proof, risk reduction, and delivery governance. We offered a short technical deep dive with their internal IT lead, not as a flex, but as a trust-building move. We shared a real story where we discovered ugly data truth early and adjusted before it became a headline. The tone changed too. Less "Here's what we can build." More "Here's how we'll make sure it doesn't become your next headache." We won the project. But the bigger win was what we learned about ourselves. ## The uncomfortable truth: AI makes it easier to be average. If we'd followed the AI output, we would have sent a perfectly reasonable follow-up. It would have sounded professional. It would have ticked the boxes. And we probably would have lost. Not because the email was bad. Because it was generic. That's the thing leaders need to get clear on: when AI commoditises answers, "good" becomes table stakes. The differentiator shifts up the stack. To judgment. To context. To the ability to challenge what sounds plausible and ask, "What assumption is hiding in here?" Critical thinking is your competitive moat. Not as a nice idea. As an operational advantage. ## What critical thinking looked like in that moment. It wasn't a philosophy discussion. It was practical. Here's what we actually did that changed the outcome: - **We treated the AI output as a draft, not a decision.** Helpful starting point. Not truth. - **We interrogated the assumptions.** What is this plan assuming the client cares about most? Is that actually true? - **We pulled in lived experience early.** The person who has seen this movie before will spot the twist. - **We asked a human question AI couldn't.** "What happened last time?" isn't a prompt trick. It's empathy and curiosity in a business suit. - **We acted decisively once the context changed.** Critical thinking isn't endless analysis. It's the courage to change course. That last point matters. Because plenty of teams spot the issue and still do nothing. They stick with the safe plan. They send the generic email. They avoid the awkward conversation. That's not a capability gap. That's an intellectual courage gap. ## Human judgment is becoming premium (and you can build it) If you're leading a sales, marketing, or customer team right now, you don't need more AI usage. You need better thinking around AI usage. That means building a culture where people are rewarded for challenging outputs, not just producing them faster. Where someone can say, "I don't buy it," without being treated like they're slowing things down. Where you make space for the inconvenient question. Because the future isn't AI versus humans. It's humans who apply judgment versus humans who outsource it. And the organisations that win won't be the ones with the best prompts. They'll be the ones with the best judgment. ## Action step: run the "AI Interrogation" habit this week Pick one AI-generated output your team is about to ship - an email sequence, a campaign plan, a proposal, a customer response. Then ask three questions in a 10-minute huddle: - **What is this assuming?** - **What could be contextually wrong even if the facts are right?** - **Who in our organisation has lived experience that should challenge this?** Do that consistently and you'll build the only advantage that won't get commoditised. The moat isn't the model. It's the mind behind it. ## Power up your potential with practical AI skills. Contact us to discuss how Artificial Intelligence could boost your business. [Book an AI Chat](https://www.newzealand.ai/contact-us/?service=ai-chat#book) [Explore the AI Operating System](https://www.newzealand.ai/ai-os/) --- # AI use cases are a distraction. Here's what we did instead Source: https://www.newzealand.ai/use-cases-distraction/ Jan, 2026 · Leadership # "AI use cases" are a distraction. Here's what we did instead. Here's the question: Which number are you willing to be held accountable for moving this quarter? **Justin Flitter** · Founder, AI New Zealand · Jan, 2026 · 5 min read I'm the CFO of a residential construction business in New Zealand. I used to think "AI strategy" meant finding clever use cases. You know the type: "Let's use AI for estimating." "Let's use AI for health and safety." "Let's use AI for marketing." It's a grab bag of good ideas, usually delivered with a slick demo and a slide that says *game-changer*. Then I looked at our month-end. Again. We're a residential construction business. Good people, solid reputation, plenty of work. But the finance reality is less glossy: cash gets tight, jobs drift, rework bites, and the margin we thought we had quietly evaporates somewhere between the quote and the final invoice. That's the moment I stopped caring about use cases. I started caring about numbers. ## The problem: We weren't short on work. We were short on control. If you've ever sat in the CFO seat of a construction firm, you'll know the feeling. Revenue looks fine on paper, but operational noise eats profit in a dozen small ways: - Variations get raised late or not at all: - Subbies show up at the wrong time, or not at all - Site notes sit in someone's phone, not in the system - The same question gets answered six times by six people - A job "mostly" finishes, then drags on for weeks - We burn hours chasing information we already have, somewhere None of that is solved by a shiny "AI use case library". That's just new vocabulary for old chaos. So I asked a different question in our leadership meeting: **"Which number are we willing to be held accountable for moving in the next 90 days?"** Silence. Then some uncomfortable shifting. Good. That's where progress starts. ## The shift: From "Where can we use AI?" to "Which lever matters most?" We picked two value levers. Not ten. Two. **Cost-to-serve (labour and admin time per build)** **Cycle time (days from contract to handover)** Because in residential construction, those two levers quietly control everything else: - Margin - Cashflow - Capacity - Customer experience - Stress levels, including mine Use cases describe what technology can do. Value levers define what the business must improve. Once we agreed on the lever, the conversation changed overnight. No more AI theatre. No more "pilot this, trial that." We weren't buying software. We were buying time and reducing leakage. ## The story: Our "AI initiative" started with one painful Friday It was 4:45pm. I was about to shut my laptop when our build manager called. "We've got a claim from a client. They're saying we didn't tell them about a material substitution. They want a discount." I asked the obvious CFO question: "What does the paperwork say?" That's when we realised the truth. The "paperwork" was split across: - site photos and texts - an email thread with the supplier - a Teams chat - a variation draft sitting in someone's inbox - notes from a site meeting that never made it into the job file We weren't losing money because we lacked effort. We were losing money because our information was scattered. And scattered information is expensive. So instead of doing what we usually do (have a post-mortem, blame the process, promise we'll do better), I wrote two lines on the whiteboard the following Monday: - **Lever: Reduce cost-to-serve** - **Metric: Admin hours per active build** Then I asked: "What's driving this number up?" Not hypotheticals. Real friction. The answer was immediate: - Our project coordinators were drowning in emails and follow-ups - PMs were rewriting the same updates for clients, suppliers, and internal teams - Variations were inconsistent and slow because they relied on perfect documentation - Everyone was duplicating work because nobody trusted where the latest info lived That's not an AI use case. That's a value leak. ## What we did: One lever, one workflow, one measurable target We didn't start with "let's implement AI". We started with a workflow we could measure: **variation management and client updates**. **We set a 90-day target:** - Reduce admin time per build by **10%** - Cut average variation turnaround time from **5 days to 2** - Improve "client update consistency" (yes, we actually measured it by audit sampling) **Then we built a simple process:** ## 1 Capture the raw reality, automatically Site supervisors already take photos and notes. The difference was this: we created a standard channel where those inputs land, tagged to the job. No heroics. Just consistency. ## 2 Use AI to turn chaos into a first draft We used an AI assistant to: - summarise site notes into structured variation drafts - pull key details (date, location, materials, reason, impact) - draft a client update email in our tone of voice, based on the latest job notes Important point: **AI didn't approve anything.** It prepared the work so humans could apply judgement quickly. ## 3 Add human judgement where it matters PMs reviewed and edited every variation and client update before it went out. That's non-negotiable. AI commoditises answers. In construction, context is everything. Critical thinking is your competitive moat. Especially when a tool is confidently wrong. ## 4 Measure the lever weekly, not quarterly If you want AI value fast, you measure fast. We tracked: - admin hours logged by role and job stage - time from "issue identified" to "variation sent" - time from "variation sent" to "variation approved" - rework incidents tied to communication gaps No vanity metrics. No "number of prompts used". Just business numbers. ## The results: Not miraculous. Real. Here's what changed within three months: - Variation drafts stopped sitting in limbo because someone "didn't have time to write it up" - Client updates became faster and more consistent, which reduced escalations - PMs spent less time writing and more time managing the job - Admin staff stopped being the organisation's human search engine - We tightened our cashflow because variations were issued earlier and approved faster Did we suddenly become a futuristic AI-enabled construction company? No. We became a slightly more disciplined business that moved two numbers in the right direction. And that's the whole point. AI didn't transform us. It removed friction. ## The lesson: Use cases are easy to collect. Value is harder. Use cases are comforting because they feel productive. You can brainstorm them all day and still avoid the hard work of choosing what matters. Value levers force decision-making: - Which metric? - What baseline? - What target? - Who owns it? - How will we measure it? What changes next Monday? When you anchor AI to a lever, you don't "pilot AI". You move a number that matters. ## If you're a CFO (or any leader), start here If I could go back and give myself a cheat sheet, it would be this: - **Pick 1–2 levers only.** (Revenue, cost-to-serve, cycle time, risk, customer experience.) **- Name the metric and baseline.** If you can't measure it, it's a hobby. **- Set a 90-day target.** Long enough to matter. Short enough to focus. **- Choose one workflow where that lever leaks value.** **- Use AI to compress the boring parts.** Drafting, summarising, extracting, standardising. **- Keep humans in the loop for judgement.** Especially in high-trust industries. Curiosity is the antidote to obsolescence. But curiosity without accountability is just tinkering. So here's the question I'm asking you the same way I asked our team: **Which number are you willing to be held accountable for moving this quarter?** ## Power up your potential with practical AI skills. Contact us to discuss how Artificial Intelligence could boost your business. [Book an AI Chat](https://www.newzealand.ai/contact-us/?service=ai-chat#book) [Explore the AI Operating System](https://www.newzealand.ai/ai-os/) --- # AI doesn't fix a broken process. It just accelerates it. Source: https://www.newzealand.ai/broken-process/ Jan, 2026 · Operating Logic # AI doesn't fix a broken process. It just accelerates it. If your business senses the wrong signals, thinks in the wrong places, decides too slowly, and acts without feedback, AI simply makes you faster at doing the wrong thing. **Justin Flitter** · Founder, AI New Zealand · Jan, 2026 · 6 min read I used to think "AI transformation" meant buying a tool, running a pilot, and waiting for the ROI graph to politely climb to the right. That's what the sales decks implied, anyway. I'm the GM of Operations at a mid-sized manufacturing business in New Zealand. We make components that end up inside other people's products, which means our customers don't care about our internal dramas. They care about delivery dates, quality, and whether we can handle the next urgent change without falling over. And if I'm honest, for a while there, we were falling over. Regularly. ## The problem wasn't information. It was how we turned it into action. We weren't short on data. We were drowning in it. - Orders came through email, calls, and the odd spreadsheet someone "just quickly updated". - Production ran off an ERP system that was fine, as long as nothing unexpected happened. Which it did. Daily. - Quality data existed, somewhere, but it took a mini-archaeological dig to find the right batch notes. - Our most accurate forecasting tool was still "Dave's gut feel", and Dave was two years from retirement. We tried bolting AI on top of that mess. We bought a shiny forecasting add-on. It looked smart in demos. It was... less impressive in the real world. Because the real world included missing fields, inconsistent part names, late updates, and a dozen micro-decisions happening in people's heads, not systems. That's when I learned the hard lesson: **AI doesn't fix a broken operating logic. It just accelerates it.** If your business senses the wrong signals, thinks in the wrong places, decides too slowly, and acts without feedback, AI simply makes you faster at doing the wrong thing. AI theatre. Better screenshots. Same bottlenecks. ## The turning point: one late shipment, one brutal customer call. The moment it really clicked was a Thursday afternoon in winter. We had a key customer waiting on a shipment. Not a "nice to have" order, a line-stopping part. We had *told* them it would be out the door by midday. At 1:30pm I got the call. They weren't yelling. That was the worst part. They were calm in that "we're updating our supplier list" kind of way. I walked onto the floor and did what I always do in those moments: I started asking questions. - Where is the job up to? - What's holding it up? - Who owns the decision to swap priorities when a hot order lands? - Why didn't we know this earlier? The answers were painful. Everyone was doing their best. They were also operating inside an invisible system of assumptions: - Planning will catch that. - Stores will tell us if we're short. - Quality will flag it if it's off. - The supervisor will make the call. Except "planning" was one person juggling six fires. "Stores" found out too late. "Quality" found it, but didn't know who to tell. And the supervisor didn't want to override anyone without backup, because last time they did, it became a political mess. It wasn't incompetence. It was **operating logic**. The system that determines how we **sense, think, decide, and act**, whether we've made it explicit or not. Ours was basically "hope and heroics". ## The solution: we stopped asking "what AI tool should we buy. Instead, we asked a better question: **How do we want this business to sense, think, decide, and act when AI is in the loop?** We picked one high-stakes workflow to redesign: *order changes and urgent jobs*. The stuff that wrecked our week. We ran a workshop with production, planning, quality, stores, and customer service. I kept it simple and a bit blunt. On the whiteboard, I wrote four headings: - **Sense**, what signals do we need early? - **Think**, what interpretation, rules, and judgement gets applied? - **Decide**, who decides, based on what criteria, and how fast? - **Act**, what happens next, and how do we learn if it worked? At first everyone wanted to jump to solutions. I didn't let them. We mapped the current reality. And it was ugly, but also clarifying. We could finally see where work went to disappear. Then we redesigned it. **Sense: we created one source of truth for "what just changed"** We stopped letting change requests live in inboxes. Every change request became a structured entry: order number, part, quantity delta, required date, customer criticality, and "why". No drama. No novel-length emails. Just the signal. **Think: we made judgement visible (and teachable)** This was the magic bit. We documented the rules we were *already* using in our heads: - What counts as line-stopping? - What lead times are real, not theoretical? - What substitutions are safe? - What jobs must not be interrupted once started? Then we built a lightweight AI assistant to summarise each change request, pull the relevant history (previous issues with that customer/part), and suggest options. Not decisions. Options. Because humans still own judgement. **Human judgement becomes premium** when AI commoditises answers. **Decide: we changed decision rights (and it was uncomfortable)** Previously, decisions happened through a game of organisational ping-pong. Now, we set a rule: urgent job triage happens at 10:00am and 2:00pm daily. Fifteen minutes. Standing meeting. The decision-maker is the shift manager *with* planning and quality present. Customer service is in the loop. If it impacts a key account, sales can join, but they don't get veto power just because they're loud. Most importantly: we wrote down escalation paths. If something is above a threshold (cost, risk, customer impact), it escalates to me. Otherwise, it gets decided on the spot. No more waiting for permission. No more "I thought someone else was handling it." **Act: we built feedback into the workflow, not into someone's memory** Every urgent change now has a simple close-out: - What did we decide? - Did it ship on time? - Did it create a defect, rework, or downtime? - What would we do differently next time? That feedback loop feeds the knowledge base the AI assistant draws from. The system gets smarter because we designed it to learn, not because the model is magical. ## What changed wasn't the tool. It was the operating logic. Within a month, the difference was obvious. Not because we suddenly became a "high-performing AI organisation". But because: - People trusted the process. - Decisions happened faster. - Everyone could see the same information at the same time. - The business stopped relying on heroics. And here's the part that surprised me: morale improved. When operating logic is unclear, people feel like they're failing even when they're working hard. When it's explicit, they can win. They can improve it. They can teach it to new starters. They can scale it. AI didn't replace our team. It made the best parts of our team, experience, judgement, context, **repeatable**. ## Why it matters: AI is a force multiplier, not a rescue boat. If your operating logic is messy, AI multiplies the mess. If it's clear, measurable, and designed for learning, AI becomes infrastructure. That's the difference between: - a pilot that looks good in a slide deck, and - an operating model that actually runs better on Monday morning. Most businesses don't need "more AI use cases". They need to decide, deliberately, **how decisions get made when AI is in the room**. Because the real strategy question isn't "Which tool should we buy?" It's: **"How do we want this business to sense, think, decide, and act in an AI-rich world?"** ## Action steps: if you're leading ops, do this next week. **Pick one workflow that regularly causes pain** - **Choose something with real consequences:** delivery slippage, quality escapes, overtime blowouts, customer churn. **Map the Sense–Think–Decide–Act loop with the people doing the work** - Not managers guessing. The actual humans in the loop. **Make decision rights explicit** - Who decides what, how fast, with what information, and what gets escalated. **Add AI only after the logic is clear** - Use AI to summarise, retrieve context, and propose options. Keep humans accountable for judgement. **Build the feedback loop** - If you're not capturing outcomes, you're not learning. If you're not learning, you're doing AI theatre. If you want a simple template, Justin can share the one he uses for mapping Sense–Think–Decide–Act, and the guardrails that stop "smart suggestions" turning into dumb automation. ## Power up your potential with practical AI skills. Contact us to discuss how Artificial Intelligence could boost your business. [Book an AI Chat](https://www.newzealand.ai/contact-us/?service=ai-chat#book) [Explore the AI Operating System](https://www.newzealand.ai/ai-os/) --- # CEO Briefing: The 2026 AI Imperative Source: https://www.newzealand.ai/2026-ceo-ai-briefing/ Nov 28, 2025 · CEO Briefing # The 2026 AI Imperative. In 2026, AI capability will define your relevance. Without it, New Zealand organisations will struggle to remain viable to competitors, customers, or employees. **Justin Flitter** · Founder, AI New Zealand · Nov 28, 2025 · 6 min read ## Executive Summary The era of incremental improvement is over. With the arrival of autonomous AI Agents, businesses are no longer competing on headcount or human hustle, but on "compute" and capability. When a competitor can execute core business functions (like auditing, compliance, or customer service) 10x faster and cheaper than you, traditional competitive strategies, hiring more people or working longer hours, become mathematically impossible to sustain. The only viable response is to transition from a "human-only" workforce to a hybrid workforce where AI agents handle the high-volume, repetitive execution, freeing humans to focus on strategy, judgment, and relationships. ## The Shift from Consumption to Contribution The narrative of Artificial Intelligence in New Zealand has shifted fundamentally. In 2025, the conversation moved from the fear of job displacement to curiosity, leaders asked, *"How do we get started?"* and teams began experimenting with tools, governance, and pilot programs. As we look toward 2026, the experimentation phase is closing. AI is becoming table stakes. The latest data reveals that 82–87% of New Zealand businesses are now using AI, with 93% reporting improved worker productivity. We are no longer talking about simple chatbots; we are entering the era of autonomous AI agents capable of executing complex, multi-step workflows across finance, law, and operations without constant human oversight. In 2026, AI capability will simply define your relevance. Without it, New Zealand organisations will struggle to remain viable to competitors, customers, or talent. ## Why This Matters to NZ CEOs For New Zealand business leaders, this is a competitiveness issue, not just a technology one. We are outpacing our neighbors: 70% of Kiwi CEOs report AI has made their workforce more efficient, compared to just 42% in Australia. This gives NZ firms a productivity edge in the region. The "Sinking Lid" on hiring: While only 7% of firms report direct job displacement, 40% report a reduced need for new hires. AI is quietly reshaping workforce demand, allowing companies to grow output without proportionally growing headcount. Trust is the new currency: Despite high business adoption, consumer trust lags significantly, only 34% of Kiwis trust AI. Closing this gap is a critical leadership challenge. Regulatory Advantage: New Zealand's "light touch, risk-based" regulatory environment offers a unique sandbox for innovation, allowing local firms to move faster than competitors in more restrictive jurisdictions like the EU. ## The Opportunity This Presents Transitioning from "users" of tools to "leaders" of agents offers three distinct strategic advantages: - **Operational Excellence via Agents:** We are moving beyond drafting emails to deploying autonomous agents that can audit financial transactions 24/7 or manage supply chains. Early adopters in NZ are seeing 74% reductions in audit efforts and 63% cuts in compliance downtime. - **Becoming a "Frontier Firm":** AI allows NZ businesses to overcome our traditional scale disadvantages. Small teams can now output enterprise-level work, creating competitive moats that large global incumbents cannot easily replicate. - **Cost-Effective Transformation:** You do not need to build from scratch. 72% of NZ businesses rely on off-the-shelf solutions (like Microsoft Copilot or ChatGPT), keeping setup costs often under $5,000. The barrier to entry is low; the barrier to mastery is culture. ## The Impact on Risk and Governance The shift to autonomous agents introduces risks that traditional governance frameworks are ill-equipped to handle. - **Autonomous Action Risk:** Unlike a tool that waits for a command, an AI agent can independently make decisions, commit resources, or interact with customers. If an agent creates a liability, who is responsible? - **The "Shadow AI" Challenge:** With widespread adoption, employees are likely using AI tools informally. Without a formal "embedded compliance" framework, you risk data leakage and reputational damage. - **Cultural Risk:** Success is 10% technology and 90% people. If adoption is limited to a few tech enthusiasts rather than being culturally embedded, you will fail to realize ROI. There is a real risk of a "two-speed" workforce where those without AI skills become obsolete. ## Questions CEOs Should Be Asking Use these questions to provoke debate with your Board and Executive Team: - "If AI agents could handle 60% of our core workflows, what would we stop doing, and what new value would we create?" (Move the conversation from efficiency to business model innovation). - "Do we have a governance framework for autonomous decisions, or just for human ones?" (Traditional approval chains break down when software acts independently). - "Are we treating AI as a tool for individuals, or as a 'team member' embedded in our processes?" (ROI comes from systemic integration, not ad-hoc usage - "How are we measuring the 'sinking lid' effect on our talent pipeline, and are we upskilling our people fast enough to fill the new high-value roles created?" - "Is our trust gap with customers growing or shrinking as we automate?" (Transparency is essential to overcome the 34% consumer trust figure). ## Steps CEOs Should Take Next **Immediate Actions (Next 30 Days):** - **Conduct a "Shadow AI" Audit:** Survey your teams to find out what tools are already being used. You cannot govern what you cannot see. - **Establish an AI Council:** Form a cross-functional group (Legal, HR, IT, Ops) to own the AI strategy. Don't leave this solely to the CTO. - **Identify 3 "Agent" Pilots:** Select three internal, low-risk processes (e.g., internal reporting, initial candidate screening, invoice processing) to test autonomous agents, not just chatbo **Medium-Term Actions (Next 6 Months):** - **Embed "Human-in-the-Loop" Governance:** Update risk frameworks to ensure human oversight exists for all autonomous agent outcomes. - **Invest in "AI Ambassadors":** Identify and train power users within every business unit to champion culture change. Remember, the barrier is usually people, not tech. - **Redefine Roles:** Rewrite job descriptions to include AI collaboration as a core competency. Make AI proficiency a KPI for performance reviews to signal that this is now "table stakes." ## Key Insights **The "10x Gap" is Structural, Not Temporary** We are shifting from "AI as a Tool" (chatbots assisting humans) to "AI as an Agent" (software doing the work independently). Traditional Firm: 10 staff members take 5 days to audit 100 files. AI-Enabled Firm: 1 staff member + 1 AI Agent takes 4 hours to audit the same 100 files. The Result: The competitor is not just "slightly" faster; they are operating at a different order of magnitude. They can lower prices, increase margins, or scale service levels in ways you cannot physically match with human labor alone. **You Cannot "Hire" Your Way Out of This** In the past, if a competitor moved faster, you could hire more staff to catch up. This logic no longer holds. The Cost Asymmetry: Adding human headcount adds linear cost and complexity. Adding AI agents adds exponential capacity at near-zero marginal cost. The "Sinking Lid": Smart companies are already enforcing a "sinking lid" on administrative hiring, using attrition to replace back-office roles with automated agents. If you continue to hire for these roles, you are baking in structural inefficiency that will make your business uncompetitive within 18–24 months. **The Only Way to Compete: "Adopt or Die"** To survive against a 10x competitor, you must adopt the same leverage points. Internal Efficiency: You must deploy agents to handle your "operational debt", the low-value, repetitive tasks that clog up your teams (e.g., inbox management, invoice matching, meeting scheduling). External Speed: Your customer response times must move from "business days" to "seconds." If a competitor's AI agent answers a query instantly at 2 AM, your 9-to-5 support team is already obsolete. **The Human Advantage: Where You *Can* Win** If AI wins on speed and volume, humans win on trust and nuance. Differentiation: When everyone has 10x speed, the competitive differentiator shifts to experience and trust. Use the time saved by AI to double down on high-touch, human relationships that AI cannot replicate. The New Skillset: Your team's value is no longer "doing the work"; it is managing the agents that do the work. The most valuable employees will be those who can orchestrate AI workflows effectively. ## Actionable Takeaways for Leaders Identify "10x" Pilots: Don't use AI for small 10% gains (like writing emails). Look for processes where you can achieve a 10x reduction in time/cost (e.g., automating the entire accounts payable process). Stop Hiring for "Robot Work": Review every open role. If the job description is primarily processing data or moving files, cancel the hire and build an agent instead. ## Power up your potential with practical AI skills. Contact us to discuss how Artificial Intelligence could boost your business. [Book an AI Chat](https://www.newzealand.ai/contact-us/?service=ai-chat#book) [Explore the AI Operating System](https://www.newzealand.ai/ai-os/) --- # From AI theatre to real impact: an NZ finance story Source: https://www.newzealand.ai/ai-theatre/ Jan, 2026 · Leadership # From AI Theatre to Real Impact: A NZ Financial Services Story. The real question wasn't, "Should we use AI?" It was, "Are we the kind of organisation that can learn fast enough to use it well?" **Justin Flitter** · Founder, AI New Zealand · Jan, 2026 · 6 min read I'm the CEO of a mid-sized financial services business in New Zealand. Justin asked if we could share our story, I asked to remain anonymous. We're not a startup. We're not a legacy dinosaur either. We're that familiar Kiwi middle ground: regulated, responsible, customer-first, and quietly allergic to risk. Which is exactly why AI landed on our doorstep like an overconfident salesperson. Everyone had an opinion. Some people wanted to buy *everything*. Others wanted to ban it from the building like it was an invasive species. Both reactions missed the point. The real question wasn't, "Should we use AI?" It was, "Are we the kind of organisation that can learn fast enough to use it well?" ## The problem wasn't the tech. It was our thinking Our first AI conversations were classic "AI theatre". A vendor demo here. A glossy slide deck there. Lots of talk about transformation. Not much change in how work actually happened. We were surrounded by impressive outputs and still unsure what to trust. And in financial services, "unsure" is just another word for "no". Here's the uncomfortable truth: most organisations don't struggle with AI because they lack tools. They struggle because they haven't built the human habits required to use the tools responsibly. Curiosity is the antidote to obsolescence. But curiosity without judgement is just expensive experimentation. That's where critical thinking earns its keep. ## The moment it got real A customer emailed in with a complaint about fees. The kind of message that arrives already hot and stays hot until someone deals with it. We had a backlog. Our service team was stretched. We were trying to keep response times down without burning people out. A well-meaning manager suggested we trial an AI tool to draft replies. It sounded sensible. It also sounded dangerous. So we ran a controlled experiment. We took 50 real complaint emails and asked the AI to draft responses using our publicly available policies and a brief on our tone of voice. Then we sat down with three people: - A senior customer service rep who knows what upset customers actually need to hear - A compliance lead who knows what we can and can't say - A product manager who understands the fee structures and edge cases We didn't ask, "Is this good?" We asked, "Where is this wrong?" That question changed everything. The AI drafts were... surprisingly decent. Warm tone. Clear structure. Confident language. And that was the problem. The confidence. In three of the 50 emails, the draft promised outcomes we couldn't guarantee. In a couple more, it used language that would be fine in casual conversation but risky in a regulated environment. In one case, it subtly misunderstood the customer's situation and offered an explanation that would have escalated the complaint. It wasn't malicious. It was just confidently wrong. That day, my view of AI snapped into focus. AI commoditises answers. Human judgement becomes premium. ## Curiosity built the capability Once we stopped treating AI like magic and started treating it like a junior analyst with a lot of confidence and not enough context, our approach matured quickly. We got curious, but in a disciplined way. We started small: - We created a simple "AI usage guideline" written in plain English, not legalese - We trained teams on how to write good briefs (context in, quality out) - We set up a fortnightly show-and-tell where staff shared wins *and* failures - We tracked time saved, error rates, and customer sentiment, not vibes Within six weeks, we weren't "doing AI". We were building an organisation that learns. The capability grew fast. Not because AI was amazing. Because our people were. And importantly, curiosity didn't just live in innovation meetings. It showed up in everyday work: - "What's the real bottleneck here?" - "What assumptions are we making?" - "What would happen if we tried this with one team for two weeks?" Those are curiosity questions. They're also business questions. ## Critical thinking deployed it with impact Then came the bigger test. Our credit team wanted to use AI to help summarise customer documents for internal review. On paper, it was a no-brainer. The time savings looked massive. The team was excited. Leadership was tempted to approve it quickly. I pushed pause. Not because I'm anti-AI. Because I've been around long enough to know that "massive time savings" is how organisations talk themselves into avoidable risk. So we ran it through a critical thinking filter. Three questions: - **What's the decision being made, and what's the risk if it's wrong?** Summaries would influence lending decisions. Wrong summary, wrong decision. High stakes. - **What data is involved, and where does it go?** Customer documents contain sensitive personal information. We needed clarity on storage, retention, access, and vendor terms. - **What's the minimum viable version of this that still adds value?** Instead of "AI summarises everything," we started with "AI drafts a summary that a human must verify, with mandatory source quotes attached." That last part mattered. We required the AI output to include citations: "This sentence comes from this paragraph in this document." No citation, no trust. That single design choice turned AI from a shortcut into a thinking partner. We also added a red-team step where someone tried to break the process: upload messy documents, contradictory info, missing pages. We assumed reality would be worse than the demo. Reality always is. When we finally rolled it out, the impact was real: faster reviews, less cognitive load, better consistency, and no spike in errors. Not because the AI was perfect. Because the system around it was. That's critical thinking. Not cynicism. Not slowing down for the sake of it. Just judgement, applied with intellectual courage. ## What I learned as a CEO AI didn't change our business overnight. It changed how we think about improvement. Curiosity made our team willing to explore. It gave people permission to learn in public, to ask "dumb" questions, and to run small experiments without fear of looking behind the curve. Critical thinking made those experiments safe, useful, and scalable. It forced us to interrogate outputs, challenge assumptions, and build guardrails that match the reality of regulated work. One without the other is a problem. Curiosity without critical thinking becomes chaos. Critical thinking without curiosity becomes stagnation. Together, they build something rare. A competitive moat. ## Action steps: if you lead a team, start here If you're a leader in New Zealand financial services (or any regulated industry), don't start by asking, "What AI tool should we buy?" Start with this: - **Ask better questions in meetings**. Swap "Do you understand?" for "What are you curious about?" once a week. Watch what changes. - **Pilot one workflow, not a whole strategy**. Pick one real process that's causing frustration. Run a two-week experiment. Measure time saved and risk introduced. - **Make human judgement non-negotiable**. Require review, verification, and context checks. Treat AI as a draft machine, not an authority. - **Reward the people who spot what's wrong**. The person who says, "This output looks confident but it's missing context" is saving you money and reputation. Celebrate that. Curiosity is the antidote to obsolescence. Critical thinking is your competitive moat. And in a world where AI commoditises answers, the winners won't be the organisations with the most tools. They'll be the ones with the best judgement. Curiosity builds capability. Critical thinking deploys it with impact. That's the play. ## Power up your potential with practical AI skills. Contact us to discuss how Artificial Intelligence could boost your business. [Book an AI Chat](https://www.newzealand.ai/contact-us/?service=ai-chat#book) [Explore the AI Operating System](https://www.newzealand.ai/ai-os/) --- # Why AI needs culture to thrive. Source: https://www.newzealand.ai/ai-needs-people-culture/ Nov 11, 2025 · People & Culture # AI for Business: Why the tech needs people & culture to thrive. Discover why successful AI adoption in business is 90% about people and culture, not just technology. **Justin Flitter** · Founder, AI New Zealand · Nov 11, 2025 · 2 min read As New Zealand surges ahead in AI adoption, business leaders are discovering that embedding artificial intelligence into their operations is 10% technology and 90% people and culture. The real challenge isn't buying the next tool, it's reimagining how work gets done and enabling every employee to confidently adopt AI in their daily workflows. Recent stats paint a remarkable picture: 87% of NZ organisations now use some form of AI, up from 48% just two years ago. But dig deeper and the business impact is scattered, often limited to experiments led by early adopters, leaving most of the ROI unrealised. True returns only show up when AI isn't just a pilot, it's embedded, it's routine, and everyone is on board. ## Reimagining Everyday Work The fast-movers aren't just automating tasks; they're rethinking whole processes. AI is being woven into document creation, customer service, compliance, and decision-making. These organisations treat AI as a "team member", redesigning workflows so humans and AI can collaborate, unlocking productivity gains that competitors can't easily copy. Yet, for most businesses the journey to embed AI remains unfinished. It takes more than workshops or toolkits. It takes leadership to inspire confidence, upskilling to build capability, and governance that enables innovation while maintaining trust and integrity. ## The Culture Gap and the Trust Challenge People and culture are the true battlegrounds. While 69% of NZ consumers use AI, only 34% feel they can trust it. In retail, just a third of workers use AI compared to over 90% in Singapore. This isn't apathy, it's a call for clarity and evidence of everyday value. Kiwis are quick adopters when AI is transparent and directly helpful, like in customer support. But where AI feels opaque or uncertain, such as product recommendations, adoption lags. ## Moving From Experiments to Everyday The lesson is clear: business-wide ROI only emerges when AI moves beyond scattered use. Leaders need to invest in practical skills training, champion AI ambassadors within teams, and create workflows where using AI is as normal as checking email. Bring everyone along on the journey, not just the tech enthusiasts. The future of work in New Zealand is being rewritten right now. And the businesses that thrive will be those with the vision, culture, and courage to make AI part of everyone's job, every day. ## Power up your potential with practical AI skills. Contact us to discuss how Artificial Intelligence could boost your business. [Book an AI Chat](https://www.newzealand.ai/contact-us/?service=ai-chat#book) [Explore the AI Operating System](https://www.newzealand.ai/ai-os/) --- # New Zealand's AI Adoption in 2025 Source: https://www.newzealand.ai/new-zealand-ai-adoption-in-2025/ Sept 29, 2025 · Industry insights # New Zealand's AI Adoption: A Comprehensive Analysis in 2025. Explore the trends in business integration, workforce changes, consumer trust, and policy. **Justin Flitter** · Founder, AI New Zealand · Sept 29, 2025 · 14 min read ## The Shift from Consumption to Contribution **Key Highlights:** - New Zealand's AI adoption surged in 2025, with 82-87% of businesses and 69% of consumers using AI tools for productivity, though trust remains a challenge. Major impacts include workforce augmentation (not replacement), strong operational efficiency gains (93-96% workers more efficient), and positive financial returns. - The nation favors off-the-shelf AI solutions over custom builds, driving rapid deployment, especially among large enterprises. - Consumer trust is lagging, only 34% trust AI. Regulatory gaps exist, but recent government strategies, increased investment, and coordinated industry and academic initiatives are steadily boosting public sector adoption. Māori engagement is progressing but still underrepresented. - **Key sectors leading adoption:** transport, media, tech, and public services. Influential contributors are the NewZealand.AI, AI Forum NZ, NZTech, AUT Centre for AI Research, university institutes, Spark, Xero, Air NZ, and various government agencies. The article highlights actionable insights and strategic challenges around skills development, trust-building, infrastructure investment, and regulatory clarity. New Zealand is positioned for future AI success if it maintains momentum in business, government, and workforce transformation while addressing the trust and capability gaps. New Zealand has undergone a remarkable transformation in artificial intelligence adoption throughout 2025, emerging as a surprising leader in business AI integration while grappling with consumer trust challenges. The country's approach to AI, prioritising adoption over development, has delivered substantial productivity gains, though questions remain about long-term competitiveness against more aggressive AI development nations. ## The State of AI in Aotearoa New Zealand's AI landscape in 2025 reveals a nation in transition. Business adoption has surged dramatically, with 82-87% of organisations now using AI in some capacity, representing nearly a doubling from 48% in 2023. I find this statistic surprising. I speak at conferences around New Zealand almost every week. I'm just not hearing evidence of AI being embedded into every day workflows as this might suggest. I clarify what these reports define as 'adoption' below. No doubt more than 80% of organisations have AI in the business tools they use everyday. Integration of AI Agents, Assistants and LLMs into daily tasks I think is much lower. However, this rapid uptake has delivered tangible results: 93% of businesses report AI has made workers more efficient, far exceeding global averages and outpacing neighbouring Australia where only 42% of CEOs report similar efficiency gains. However, this business enthusiasm contrasts sharply with consumer sentiment. While 69% of New Zealanders use AI regularly, only 34% trust the technology, and 44% believe AI risks outweigh the benefits. This trust deficit, combined with New Zealand's 40th place ranking in the Oxford Government AI Readiness Index, highlights the complex challenges facing the nation's AI ambitions. ## Business Adoption: Leading the Pack ### **Dramatic Growth Trajectory** The pace of AI adoption in New Zealand businesses has been nothing short of exceptional. From a baseline of 48% in 2023, adoption jumped to 66-67% in 2024 and reached 82-87% by 2025. Among large enterprises with over 200 employees, adoption rates are even higher at 92%, indicating that scale provides both resources and incentives for AI integration. This growth trajectory significantly outpaces international comparisons. Australia, despite its larger economy, reports only 50% of businesses regularly using AI, while New Zealand's adoption rate approaches that of Singapore's 70%+, despite Singapore's massive $27 billion AI investment and third-place global AI ranking. ### **Productivity and Financial Impact** The financial returns from AI adoption in New Zealand are substantial. 56% of businesses report positive financial impact from AI, with 71% achieving operational cost savings. These savings, combined with efficiency gains, have created a compelling business case for continued investment. Productivity improvements are particularly striking, with 93-96% of workers reporting increased efficiency due to AI. This figure represents one of the highest reported efficiency gains globally and suggests that New Zealand businesses are successfully implementing AI in ways that augment rather than replace human capabilities. ### **Implementation Preferences and Costs** New Zealand businesses demonstrate a clear preference for pragmatic AI solutions. 72-73% use off-the-shelf AI tools like ChatGPT, Microsoft Copilot, and Google Gemini, while only 13% have invested in custom-built solutions. This preference reflects both cost considerations and the rapid availability of powerful generative AI tools that require minimal technical expertise. Implementation costs have decreased significantly, with more businesses reporting setup costs under $5,000 compared to previous years. This cost reduction has democratized AI access, enabling smaller enterprises to experiment with AI without substantial upfront investment. ## Workforce Transformation: Augmentation Over Replacement ### **Minimal Job Displacement** Despite widespread fears about AI-driven job losses, New Zealand's experience suggests a more nuanced reality. Only 7% of organizations report AI directly replacing workers, a figure that has remained stable across multiple surveys. However, 40% report reduced need for new hires, indicating that AI is reshaping workforce demand rather than causing mass unemployment. This "sinking lid" approach, where businesses upskill existing employees rather than replacing departing staff, appears to be the dominant pattern. 81% of businesses now support internal and/or external AI training, demonstrating a strong commitment to workforce development. ### **New Opportunities and Skills Evolution** Paradoxically, while reducing some hiring needs, AI is simultaneously creating new opportunities. 62% of businesses report that AI is generating new career opportunities, a 13 percentage point increase from previous surveys. This suggests that AI is driving job evolution rather than elimination, creating demand for AI-literate workers across various sectors. The skills gap remains significant, however. Only 36% of New Zealanders feel they have the skills and knowledge to use AI appropriately, and just 24% have received formal or informal AI training. This skills deficit represents both a challenge ## Consumer Perspectives: The Trust Challenge ### **High Usage, Low Trust** New Zealand consumers present a fascinating paradox: high AI usage coupled with significant skepticism. 69% use AI regularly, yet only 34% are willing to trust AI. This trust deficit is more pronounced than in many other markets and represents a significant barrier to broader AI adoption. The primary concerns are practical rather than theoretical. 59% report experiencing loss of human interaction and connection due to AI, while 75% are unsure online content can be trusted as it may be AI-generated. These concerns reflect real user experiences rather than abstract fears about AI's future impact. ### **Regulatory Expectations** New Zealanders demonstrate strong appetite for AI governance. 81% believe AI regulation is required, though only 6% are aware of current AI regulations and policies. This regulatory gap contributes to uncertainty and may explain some of the trust challenges facing AI adoption. The demand for regulation is accompanied by specific expectations for accountability. 85% would be more willing to trust AI systems if there is assurance about their trustworthy use, including knowing who is accountable if issues arise. This suggests that clear governance frameworks could significantly improve public acceptance. ## Government and Public Sector Progress ### **Policy Framework Development** New Zealand launched its first national AI strategy in July 2025, making it the last OECD country to develop such a framework. The strategy focuses on adoption rather than foundational AI development, acknowledging New Zealand's resource constraints while leveraging its agility and regulatory stability. The public sector has shown rapid AI integration. The 2025 cross-agency AI survey recorded 272 AI use cases across 70 agencies, more than doubling from 108 cases across 37 agencies in 2024. 55 AI cases are now fully operational, up from just 15 in 2024, indicating genuine progress beyond pilot projects. ### **Investment and Infrastructure** Government investment in AI capabilities has increased substantially. Budget 2025 allocated $213 million for tuition and training subsidies and $64 million for STEM and priority skills, signaling recognition of the skills gap challenge. The establishment of frameworks like the Public Service AI Framework and partnerships with organisations like Callaghan Innovation demonstrate a coordinated approach to AI governance and development. ## International Comparisons: Strengths and Challenges ### **Competitive Position** New Zealand's AI adoption rates compare favourably internationally, particularly in business contexts. The 70% of New Zealand CEOs reporting AI workforce efficiency improvements versus 42% in Australia suggests superior implementation effectiveness, despite Australia's larger AI ecosystem and investment levels. However, New Zealand faces significant challenges in other areas. The 40th place ranking in the Oxford Government AI Readiness Index reflects weaknesses in government preparedness, digital infrastructure, and innovation capacity. This ranking places New Zealand well behind AI leaders like Singapore (3rd globally) and raises questions about long-term competitiveness. ### **Regional Context** Within the Asia-Pacific region, New Zealand's position is mixed. While business adoption rates are strong, the country lacks the massive infrastructure investments seen in Singapore ($27 billion) or Australia's more developed AI startup ecosystem. Singapore's market is projected to grow from $1.05 billion to $4.64 billion by 2030, while New Zealand's much smaller market may struggle to achieve similar scale. ### **Trust and Cultural Factors** New Zealanders' cautious approach to AI contrasts with more aggressive adoption in markets like India, China, or even Singapore. This cultural preference for careful, ethical implementation may slow adoption in sensitive areas but could enable more sustainable, publicly acceptable AI integration over time. ## Sector-Specific Adoption Patterns ### **Leading Industries** AI adoption varies significantly across sectors, with transport, postal and warehousing showing the highest uptake, followed by information, media and telecommunications, and public administration. Professional, scientific and technical services represent the largest single group of survey respondents at 32.6%, indicating strong adoption in knowledge work. Arts and recreation show relatively low AI uptake, possibly reflecting the sector's emphasis on human creativity and the limited applicability of current AI tools to artistic expression. ### **Use Case Distribution** The top five AI use cases remain consistent across surveys: marketing, administration, software development, design, and project management. These areas benefit from AI's strength in processing digital information and automating routine tasks while presenting lower risks than sensitive applications like healthcare or finance. Software development shows particularly high AI integration, with companies like Fraxional AI reporting 40-50% productivity improvements through AI-powered coding assistance. This aligns with global trends where development tools like GitHub Copilot have achieved widespread adoption. ## Māori Perspectives and Inclusive Development ### **Engagement Levels** The inclusion of Māori perspectives in AI development represents both progress and ongoing challenges. 39% of businesses engage with Māori during AI design and development processes, exceeding initial expectations but still leaving the majority without meaningful Māori involvement. Dr. Karaitiana Taiuru, Chair of the AI Forum's Kāhui Māori Atamai Iahiko, notes that while results are promising, most organizations still do not engage Māori at all. With the Māori economy estimated at $30 billion (8.4% of GDP) in 2023, and projections of 1.2 million Māori (20% of population) by 2040, this engagement gap represents both an ethical concern and a business opportunity. ### **Cultural and Economic Considerations** The emphasis on Māori engagement reflects New Zealand's unique constitutional framework under Te Tiriti o Waitangi. As AI becomes increasingly embedded in business and government processes, ensuring Māori voices are included in design decisions becomes crucial for both cultural appropriateness and commercial success. ## Future Outlook and Strategic Implications ### **Competitive Positioning** New Zealand faces a critical juncture in its AI journey. The country has achieved impressive adoption rates and productivity gains, demonstrating that its focus on AI application rather than development can deliver results. However, the 40th place global ranking and significant trust challenges suggest vulnerabilities that could undermine long-term competitiveness. The government's "adoption-first" strategy acknowledges economic realities, New Zealand cannot compete with the hundreds of millions invested in foundational AI development by tech giants. However, this approach requires excellence in implementation, governance, and skills development to maintain competitiveness with nations pursuing more aggressive AI strategies. **Critical Success Factors** Several factors will determine New Zealand's AI success: Skills Development: The 24% AI training rate and 36% skills confidence represent significant constraints. Massive upskilling initiatives will be essential to capitalize on AI opportunities. Trust Building: The 34% trust rate limits AI adoption in consumer-facing and sensitive applications. Clear governance frameworks and transparency measures could address these concerns. Infrastructure Investment: While current adoption relies primarily on cloud-based AI services, ensuring adequate digital infrastructure remains crucial for future competitiveness. Regulatory Clarity: The 81% demand for AI regulation coupled with 6% awareness of existing policies suggests need for clearer, more visible governance frameworks. ### **Conclusion: A Nation in Transition** New Zealand's AI story in 2025 is one of rapid business adoption, significant productivity gains, and emerging governance frameworks, tempered by consumer skepticism and international competitive pressures. The country has successfully leveraged its agility and pragmatic approach to achieve AI adoption rates that rival much larger economies. However, the path forward requires addressing fundamental challenges: building public trust, closing skills gaps, and ensuring that adoption-focused strategies don't leave New Zealand dependent on foreign AI capabilities. The next phase of New Zealand's AI journey will determine whether the country can translate early adoption success into sustained competitive advantage in an increasingly AI-driven global economy. The data from 2025 suggests New Zealand is well-positioned to succeed, but only if it can maintain momentum while addressing the underlying trust and capability challenges that could limit its AI potential. With 96% of workers reporting AI efficiency gains and a government finally committed to supportive policy frameworks, New Zealand has the foundation for AI success, now it must build upon it. ## Which organisations, companies, initiatives, or programs have contributed or influenced the adoption of AI in New Zealand over the last 10 years? The AUT Centre for Artificial Intelligence Research (CAIR), established in 2000, laid the earliest groundwork for AI research in New Zealand. Professor Albert Yeap's cognitive modeling work provided the academic foundation that would later influence practical applications. This was followed by the University of Auckland's NAOInstitute (2021), which takes a transdisciplinary approach to natural, artificial, and organisational intelligence research. The AI Researchers Association (AIRA), founded in 2021, now coordinates over 480+ researchers across all eight New Zealand universities and Crown Research Institutes, creating the academic backbone supporting today's AI developments. The University of Waikato's AI Institute (2021) and UC AI Research Group (2022) have further strengthened the research ecosystem. ### **Industry Coordination and Advocacy** The AI Forum New Zealand, founded in 2017, has been the single most influential organization driving AI adoption nationwide. Starting from what felt like a "national vacuum," the Forum has grown to 144+ member organizations and produces the critical productivity reports that track New Zealand's AI progress. Their AI Blueprint (2024) became the foundation for government policy development. NZTech and the Digital Technologies Industry Transformation Plan have provided broader industry coordination, while professional service firms like Deloitte, KPMG, and Datacom have delivered implementation support and market research. [NewZealand.AI](https://www.newzealand.ai/), founded in 2017, has established itself as a go-to resource for connecting businesses, sharing AI adoption stories, and offering advisory services tailored for New Zealand's unique market needs. - **Workforce and Business Insights:** The platform publishes regular workforce, market, and industry reports. These have shaped public understanding about skills gaps, practical use cases, and how Kiwi firms are deploying generative AI, providing real, actionable insights for business leaders. - **Events, Education & Advocacy:** Through workshops, webinars, and collaborative initiatives with groups like the AI Forum and NZTech, [NewZealand.AI](https://www.newzealand.ai/) has helped upskill thousands of business professionals. The site is often cited in government and industry policy discussions as a trusted voice urging practical action. - **Bridging Strategy and Implementation:** [NewZealand.AI](https://www.newzealand.ai/) is notable for translating global AI trends into relevant New Zealand context, making complex concepts accessible and supporting businesses from discovery to integration. [NewZealand.AI](https://www.newzealand.ai/) provides a central hub for knowledge-sharing, talent development, and practical support for the New Zealand AI ecosystem, all in ways the official government channels and academic programs simply can't replicate. ### **Corporate Pioneers and Early Adopters** Spark New Zealand began its AI journey in 2014 with machine learning experiments that evolved into the B.R.A.I.N platform by 2020. This marketing AI system achieved 10x improvement in campaign outcomes and positioned Spark as a leader in AI-driven customer experience. Their commitment extends to helping other businesses through the AI for Business Mini-MBA programme. Xero, founded in Wellington in 2006, laid crucial cloud infrastructure foundations before launching major AI integration in 2023. Their JAX AI assistant (2024) represents one of the most successful small business AI applications globally, handling everything from invoice creation to financial forecasting. Air New Zealand has been steadily building AI capabilities since launching the Oscar chatbot in 2016, progressing to computer vision AI trials in 2020, and culminating in their groundbreaking OpenAI partnership in 2025. This collaboration, deploying ChatGPT Enterprise to 3,500 employees, represents New Zealand's largest enterprise AI implementation. ### **Government Leadership and Policy Development** MBIE led the development of New Zealand's first AI Strategy launched in July 2025, making New Zealand the last OECD country to establish such a framework. The Government Chief Digital Officer (GCDO) coordinates public sector AI adoption through the Public Service AI Framework. Callaghan Innovation delivers the AI Activator program, while the newly established New Zealand Institute for Advanced Technology (NZIAT) has committed $70 million over seven years for AI research and commercialisation. Stats NZ, as Government Chief Data Steward, ensures the data infrastructure necessary for AI development, while the cross-agency AI survey tracks government adoption progress. ### **International Partnerships and Technology Access** Microsoft New Zealand has been crucial in providing enterprise AI tools and conducting economic impact research, projecting $76 billion in GDP contribution by 2038. AWS provides cloud infrastructure and enterprise AI capabilities. The Air New Zealand-OpenAI partnership represents the most significant international AI collaboration, giving New Zealand direct access to cutting-edge AI development. Tata Consultancy Services' five-year digital transformation partnership with Air New Zealand demonstrates how international expertise accelerates local AI capabilities. **Key Transformation Catalysts** Several factors created the conditions for New Zealand's AI revolution: The Government's cloud-first policy (2012, refreshed 2023) established the digital infrastructure foundation that enabled rapid AI adoption. This policy precedent demonstrated how government leadership could drive technology transformation. The COVID-19 pandemic accelerated digital transformation, creating both necessity and opportunity for AI adoption across all sectors. The availability of accessible AI tools like ChatGPT from late 2022 democratised AI access, enabling the rapid adoption rates we see today. ### **Supporting Ecosystem Development** Organisations like Simply Privacy ensure responsible AI governance, while academyEX provides training and capability development. ElementX, Capability Collective, and Bright Think offer specialised AI implementation support. The Digital Technologies Industry Transformation Plan provides strategic direction for the broader tech sector, while initiatives like Invest New Zealand create pathways for international AI investment. ### **Current State and Future Direction** This ecosystem has produced remarkable results, 82-87% business AI adoption in 2025, 96% of workers reporting efficiency gains, and positioning New Zealand ahead of much larger economies in practical AI implementation. The combination of strong academic research, coordinated industry advocacy, government policy support, and corporate leadership has created a uniquely effective AI development environment. While New Zealand may be the smallest OECD country to develop an AI strategy, the collaborative approach between these organizations has delivered adoption rates that rival global leaders. Looking ahead, the $70 million NZIAT investment, expanded university programs like AUT's new Data Science and AI Department, and continued corporate partnerships position New Zealand well for the next phase of AI development. ## Power up your potential with practical AI skills. Contact us to discuss how Artificial Intelligence could boost your business. [Book an AI Chat](https://www.newzealand.ai/contact-us/?service=ai-chat#book) [Explore the AI Operating System](https://www.newzealand.ai/ai-os/) --- # 5 Reasons Plato Would Be Skeptical About AI Source: https://www.newzealand.ai/5-reasons-plato-would-be-skeptical-about-ai/ 26 March 2025 · Gen AI # The Intersection of Technology and the Humanities **Justin Flitter** · Founder, AI New Zealand · 26 March 2025 · 5 min read When I was a young man in Santa Fe, New Mexico, I immersed myself in the Great Books of history. I studied the humanities intensely at [St. John's College](https://www.sjc.edu/about/campuses/santa-fe). This eclectic school had no tests, no grades - we read challenging books, and talked about them. Twice a week in Seminar, we engaged with the original texts of thinkers like Plato, Aristotle, and Ptolemy to work through the greatest questions posed by the Western canon. The Socratic method, questioning assumptions, and digging deeper into the nature of knowledge, was central to our method of learning. We asked questions to uncover our own ignorance. As Generative AI becomes more embedded in our lives, I find myself using the Socratic method to refine the outputs of LLMs like ChatGPT, and I can't help but wonder what Plato would say about our growing dependence on artificial intelligence. Based on his writings, I think he'd have some serious concerns. Here are 5 reasons he would have doubts: ## 1. AI Only Creates the Illusion of Wisdom In the *Phaedrus*, Plato critiques the invention of writing through Socrates, who is the star of every dialogue. He claims that writing weakens memory, and creates an illusion of understanding without true wisdom: *"You provide your students with the appearance of wisdom, not with its reality." (Phaedrus 275a)* **AI can produce text that sounds profound, but it doesn't understand what it's saying.** If Plato was worried about writing, imagine what he'd say about AI-generated content! He would argue that relying on AI gives us the illusion of intelligence, while actually weakening our ability to think critically, and stunting us from retaining knowledge. ## 2. AI Is an Imitator, Not a Thinker In Book X of the *Republic*, Plato warns about the dangers of imitation. He criticizes artists for creating representations of reality (rather than reality itself) which places them three steps removed from the Forms, the true essence of things: *"The imitator, or maker of the image, knows nothing of true existence; he knows appearances only." (Republic 601c)* **AI is the ultimate imitator.** It takes human-created content fed into it's training data, and rearranges it into something that looks original, but is merely a remix of what already exists. Plato would likely argue that AI-generated knowledge is just another shadow on the wall of the cave, and not the real thing. ## 3. AI Undermines Personal Discovery In *Meno*, Plato presents the idea of *anamnesis*, or recollection. He uses it to suggest that true knowledge is discovered within, and not externally granted. Socrates famously demonstrates this by guiding a slave boy to uncover geometric truths on his own: *"They say the soul of man is immortal, and at one time has an end, which is termed dying, and at another is born again, but never perishes... Having seen all things, whether in this world or in the world below, [the soul] has knowledge of them all." (Meno 81c-d)* **What happens when we rely on AI to provide knowledge instead of engaging with the struggle of learning?** Plato would warn us that true wisdom requires active discovery, reflection, and reasoning. Passive consumption of AI-generated answers lets AI "think" for us, and "learn" for us. If we are lazy thinkers, we risk losing the journey that makes us wise. ## 4. AI Can Be Used to Manipulate Rather Than Inform Plato warns against the power of rhetoric in *Gorgias*, because rhetoric can persuade without leading to truth. GenAI, especially with its potential for deepfakes, misinformation, and persuasive chatbot-generated content, could be a modern example of this danger: *"The rhetorician need not know the truth about things; he has only to discover some way of persuading the ignorant that he has more knowledge than those who actually have it." (Gorgias 459a-c)* **The biggest risk of AI may not be its lack of wisdom, but its ability to sound wise.** I can imagine Plato cautioning us that AI can be used to manipulate beliefs and spread convincing falsehoods, just as skilled rhetoricians could lead people astray in ancient Athens. If we aren't vigilant, we may find ourselves deceived by machines that appear insightful, but have no true understanding. ## 5. AI Lacks Moral Judgment In *Apology*, Plato presents Socrates as a man who chooses truth and morality over convenience and safety. Socrates defends his lifelong pursuit of wisdom and ethical inquiry, even at the cost of his own life. He famously declares: *"The unexamined life is not worth living." (Apology 38a)* **AI, by contrast, does not examine life.** While AI can be programmed to reflect human ethics, it doesn't understand them. Plato might say that relying on AI for decision-making could be dangerous, as it lacks the ability to reflect on justice, virtue, or the broader consequences of its outputs. If we aren't careful, we might hand over moral authority to something that doesn't even comprehend morality. The ancient Greeks had a word for skilled craftsmanship: *techne*. More than just technical knowledge, *techne* was the mastery of a craft, blending skill and wisdom in the pursuit of excellence. Athena, the goddess of wisdom and strategy, was often associated as the patron of *techne*, not just as an abstract ideal, but as a guide for applying knowledge with purpose and virtue. If AI is merely a tool, it lacks *techne*, because it does not apply skill with wisdom or intention. It can generate, but it does not create in the way a true craftsperson does. As Steve Jobs famously said: *"It is in Apple's DNA that technology alone is not enough, it's technology married with liberal arts, married with the humanities, that yields us the results that make our hearts sing."* In the end, GenAI can be virtuous when it is used as a tool that augments human wisdom, but it can be villainous when it is merely a replacement for it. The imagined critiques of Plato that I shared above all rely on AI replacing human thinking, which is not the right way to use it. When we use AI as an amplifier, and not as a replacement, we sidestep the Platonic concerns I listed above. Look, I'm not here to be a cheerleading advocate for GenAI. I won't tell you that it's great in all circumstances no matter w, no matter what. Just because it's in my logo doesn't mean I will always support it in every circumstance. I pride myself on being a skeptical thinker. I like to try on the positions of my opponents, so I can understand why they think the way they do - and to be open-minded enough to change my own position, if it gets me closer to the truth. So while many AI educators will ignore the concerns around Generative AI, I am willing to engage with these concerns directly, because I think it makes me a better teacher, and I know it makes me a better thinker. If this article has made you think a little, and you have an interesting thought to share, I'd love to know. Reply or comment, especially if you think differently on this topic than I do. ## Power up your potential with practical AI skills. Contact us to discuss how Artificial Intelligence could boost your business. [Book an AI Chat](https://www.newzealand.ai/contact-us/?service=ai-chat#book) [Explore the AI Operating System](https://www.newzealand.ai/ai-os/) --- # Anthropic's Mythos and the RBNZ warning: a New Zealand read Source: https://www.newzealand.ai/a-new-zealand-read-on-anthropic-s-mythos-the-rbnz/ 9 June 2026 · Guidelines # What actually happened **Justin Flitter** · Founder, AI New Zealand · 9 June 2026 · 5 min read For the first time, the RBNZ has named a specific AI model as a risk to the financial system. In its May 2026 Financial Stability Report, the Reserve Bank of New Zealand singled out Anthropic's frontier model **Mythos** as an example of how increasingly capable AI "could materially amplify cyber risks from malicious actors." That is not a vendor demo. That's our regulator putting a model's name in an official stability document and telling you it is watching. If you run a New Zealand business, the instinct is to file this under "cyber team's problem." That is the wrong filing. This is an operating logic problem, and it lands on the CEO's desk. Three things converged in a few weeks, and they are connected. **One. Anthropic built something it won't sell you.** Mythos is a general-purpose frontier model that, in Anthropic's own testing, found more than 10,000 high- or critical-severity vulnerabilities across major operating systems, browsers and critical software. Anthropic decided it was too dangerous for general release and instead stood up Project Glasswing, a consortium that now spans roughly 150 organisations in more than 15 countries, giving vetted defenders early access. ([Anthropic](https://www.anthropic.com/news/expanding-project-glasswing)) **Two. The RBNZ noticed.** The Financial Stability Report flagged AI on two fronts: operational resilience (Mythos-class models supercharging attackers) and credit risk (if AI drives job losses, more borrowers struggle to repay). The Bank says it is "actively monitoring the risks that Anthropic's Mythos model may pose to the New Zealand financial sector" and engaging Trans-Tasman counterparts. ([interest.co.nz](http://interest.co.nz)) **Three. The US moved to get models first.** On 2 June 2026, President Trump signed an executive order asking AI companies to voluntarily give the federal government up to 30 days of access to "covered frontier models" before wider release, with the NSA, CISA and NIST building a classified benchmark for "advanced cyber capabilities." ([CNN](https://www.cnn.com/2026/06/02/tech/executive-order-trump-ai), [Tom's Hardware](https://www.tomshardware.com/tech-industry/artificial-intelligence/trump-signs-ai-executive-order-seeking-30-day-government-access-to-frontier-models-before-release)) Read together: the most powerful cyber-capable models now go to governments and vetted critical-infrastructure players first. New Zealand has no equivalent arrangement, no seat at that table, and no domestic frontier capability of its own. ## Why this matters for a Kiwi business The capability gap is no longer theoretical. Anthropic itself warns that within 6 to 12 months, other developers will have Mythos-class models, and some may release them **without** the safeguards Anthropic chose to apply. Cheap, fast, cyber-capable models are around the corner. ([Anthropic](https://www.anthropic.com/news/expanding-project-glasswing)) So the defender's advantage that Mythos represents is temporary and gated. The attacker's version is coming, ungated. Here is the uncomfortable asymmetry for New Zealand: - The best defensive AI is being handed to the US government and global critical infrastructure first. We are downstream of that decision. - Our financial system already showed its fragility this year. A single FMI outage at the RBNZ's own ESAS disrupted around $4.5 billion of transactions and cascaded to at least three other systems before it was resolved in three hours. That was an internal glitch, not an attack. ([interest.co.nz](http://interest.co.nz)) - APRA across the Tasman has already told banks and insurers it wants "a step-change" in how they manage AI risk, with enforcement on the table. The regulatory direction is one-way. This is not a reason to panic. It is a reason to get your operating logic right before the cheap attackers arrive. ## The four decisions a CEO should make now This is where most leadership teams reach for a tool. Wrong move. AI doesn't fix a broken process, it accelerates it. The question isn't "which security product do we buy," it's how your business senses, thinks, decides and acts when AI is in the loop, on both sides of the firewall. **1. Own the cyber lever at board level.** Name one director or executive accountable for AI-related cyber resilience, with a baseline (your current patch latency, your dependency on third-party AI providers) and a 90-day target. Move a metric, not a use case. If concentration risk is your exposure (everyone leaning on one or two AI vendors), measure it and reduce it. **2. Map your real dependencies.** The RBNZ's specific worry is "relying on only a small number of third party AI providers." Most NZ firms have no idea how many critical workflows already route through one model or one cloud. Map it. One source of truth. You cannot manage a dependency you cannot see. **3. Set guardrails on what AI may and may not do.** Decide explicitly: what AI drafts, what humans approve; what AI flags, what humans decide; what must never be delegated. Write down the override and escalation paths. AI drafts. Humans approve. That sentence should be policy, not aspiration. **4. Build judgement habits, not just defences.** The risk in this era is not only hostile AI. It is your own team uncritically accepting AI output that is confidently wrong. Critical thinking is your competitive moat. Reward the person who spotted what was wrong, not just the person who shipped fast. ## A 7-day plan You can start this week. None of this requires a big budget. - **Day 1-2:** Ask your CISO or IT lead one question: "If a Mythos-class model in hostile hands targeted us tomorrow, where would it get in first?" Write down the top three answers. - **Day 3:** List every business-critical workflow that already depends on an external AI provider. Note which vendor, and what happens if that vendor goes down or is breached. - **Day 4:** Name your accountable owner for AI cyber resilience. One person. Real authority. - **Day 5:** Draft your three-line AI guardrail: what AI drafts, what it flags, what it must never decide alone. - **Day 6:** Check your cyber-security framework against the RBNZ's Cyber Resilience Guidance if you're a regulated entity, or against a recognised baseline if you're not. - **Day 7:** Brief your leadership team. Not on the technology. On the operating logic. ## The bottom line A central bank naming a model is a signal, not a scare. The capabilities that let Mythos find 10,000 critical flaws are the same capabilities that will, before long, be available cheaply to people who do not have Anthropic's caution. New Zealand will not be first in line for the defensive version. We will be very much in line for the offensive one. The firms that come through this well won't be the ones with the flashiest security tool. They'll be the ones that decided, deliberately, how their business senses, thinks, decides and acts with AI in the loop. AI commoditises answers. Human judgement becomes premium. That's not a cyber project. That's an operating system. That's the move. ## Sources - Reserve Bank of New Zealand, Financial Stability Report (May 2026), reported via [interest.co.nz](http://interest.co.nz) and [RBNZ](https://www.rbnz.govt.nz/financial-stability/financial-stability-report/financial-stability-reports/2026/may/financial-stability-report-may-2026/web-version) - [Anthropic, "Expanding Project Glasswing" (2 June 2026)](https://www.anthropic.com/news/expanding-project-glasswing) - [CNN, "Trump signs executive order asking for access to new AI models before they launch" (2 June 2026)](https://www.cnn.com/2026/06/02/tech/executive-order-trump-ai) - [Tom's Hardware, "Trump signs AI executive order seeking 30-day government access to frontier models" (2 June 2026)](https://www.tomshardware.com/tech-industry/artificial-intelligence/trump-signs-ai-executive-order-seeking-30-day-government-access-to-frontier-models-before-release) *This briefing reflects publicly reported information as at 9 June 2026. The 6 to 12 month timeline for Mythos-class models becoming widely available is Anthropic's own stated expectation, not a certainty.* ## Power up your potential with practical AI skills. Contact us to discuss how Artificial Intelligence could boost your business. [Book an AI Chat](https://www.newzealand.ai/contact-us/?service=ai-chat#book) [Explore the AI Operating System](https://www.newzealand.ai/ai-os/) --- # AI for Business: why now is the time for Kiwi SMEs Source: https://www.newzealand.ai/ai-for-business-why-now-is-the-time-for-kiwi-smes-to/ 16 November 2025 · Guidelines # AI for Business: Why Now is the Time for Kiwi SMEs to Harness the Power of Artificial Intelligence If ever there was a call to arms for New Zealand's small and medium businesses, the Business North Harbour Conference was it. **Justin Flitter** · Founder, AI New Zealand · 16 November 2025 · 5 min read The central message rang true and loud last Friday afternoon: **AI is no longer just the future, it's the now.** For leaders looking to protect, grow, and future-proof their enterprises, understanding and using AI is becoming as vital as electricity to your operation. **AI is not just about tech, it's about people, problem-solving, and culture** Justin Flitter, founder of AI New Zealand, brought the voice of practical leadership to the Business North Harbour Conference, urging New Zealand SMEs to take action by focusing on people, not just platforms. *"Don't chase every shiny tool - focus on real business problems, map your processes, and look for the bottlenecks. AI should always serve your outcomes,"* said Flitter. He highlighted how SMEs can start by using AI for tangible challenges like automating repetitive admin, transcribing meetings, and making knowledge searchable across the organization. He championed the fusion of human and machine capability: *"The best results come from workflows where people and AI collaborate. Start with human-in-the-loop pilots, listen closely to your team, and only scale what truly moves the dial."* Flitter also emphasized the game-changing importance of data: *"Your company's unique knowledge and customer insights are your new competitive edge. Capture it, structure it, and put it to work."* Flitter closed by encouraging business owners to start small, scale what works, and always prioritise trust and compliance: *"Address privacy and compliance as you grow. AI is a journey, democratise learning so every member of your team can benefit, not just the techies."* Justin's pragmatic approach gave leaders a powerful roadmap, blending safe experimentation with a growth mindset. *"What would you do with an extra five hours a week? That's the opportunity AI gives us, time to get strategic, creative, and connected with customers like never before."* **From Boardrooms to Back Offices, Real Benefits, Real Stories** This event was packed with actionable wisdom and practical insights from New Zealand's best and brightest AI trailblazers, including Greg Cross (Soul Machines & [Eighty20.AI](http://eighty20.ai/)), Dave Wilson (IT360), Matt Ensor (Kia Ora AI), Asa Cox (Arcanum AI), and more. The advice was clear: **start with your pain points, not the shiny tools.** Dave Wilson showed how using your business as a testbed beats jumping recklessly into every new platform. From meeting transcription and analysis to automating audit reports and building action lists straight from your inbox, the gains in productivity are already within reach. **"Don't chase every new tool. Focus on clarity, outcomes, and what actually moves the dial for your business,"** urged Wilson. Get your business analysts mapping processes and pinpointing bottlenecks. Clarity of objective is half the battle! **Build Hybrid Human + AI Workflows - Scale What Works** AI shouldn't just replace human skills - it should amplify it. **Start with human-in-the-loop workflows, running pilots and scaling up what actually delivers.** This approach both builds confidence internally and keeps your business resilient, adaptable, and ready for the next leap forward. Greg Cross championed treating AI as an enabler, like electricity was to the industrial age. **"AI-native strategy will reshape how Kiwis do business. LLMs are democratising innovation, so empower your AI-forward teams and avoid the trap of Shadow AI."** The lesson: harness curiosity inside your team instead of banning experimentation. **Innovation is bottom-up.** **Data - The New Oil for Kiwi Businesses** Panelists agreed: **your proprietary data is now your most valuable asset.** Capturing, structuring, and leveraging this data, whether customer interactions, operations, or internal knowledge, will determine who thrives. **Curiosity and problem-solving skills will matter even more than technical expertise.** Embracing transparency, choosing trustworthy vendors, and prioritizing data governance will set tomorrow's winners apart. "Trust is the new digital currency. Get deliberate about how you pick your vendors and make transparency a priority. You want your customers and partners to feel confident every step of the way," Cross added. **Safe Experimentation - The Cure for 'AI Fear'** Matt Ensor drove home a crucial point for Kiwi leaders: **build safe environments for staff to experiment, test, and learn.** Use internal GPTs, create AI labs, or simply set up sandboxes within your business systems. Containerise where needed and stand on the shoulders of cloud providers for data security instead of reinventing the wheel. Getting experimental lowers the risk of Shadow AI and helps business leaders move at pace without fear. **"Address compliance and insurance as a matter of course, cyber and AI risks are rising. But don't let them freeze you in place," Ensor warned.** **Practical Adoption - Not Just for the Techies** Asa Cox inspired the crowd with this promise: **you don't need to be a tech wizard to make AI seriously work for your business.** His hands-on workshops laid out no-nonsense strategies any team can follow, instantly demystifying AI and giving even the smallest shop a roadmap to automation and growth. The future isn't just for techies, it's open to every Kiwi business that wants to thrive. Start small, measure your wins, and build confidence step by step. **Panel Insights, The Questions NZ SMEs Are Asking** - *Where do we start?* Use case-driven pilots - pick a pain point and solve it! - *How do we nurture adoption, and show ROI?* Run internal workshops, model leadership, showcase real wins. - *Data & security?* Structure access and usage, leverage best-in-class cloud security, containerize where needed, and educate your staff. - *What's the one tool you can't live without?* Choose what solves the most critical problem for your business, and don't be afraid to iterate. - *What does the future hold?* In just 1-2 years, expect AI to be woven into every workflow - SMEs will have smarter, personalised customer engagement, predictive analytics, and more time freed up for creative work. **The Human-AI Advantage - Culture Eats Tech for Breakfast** Angela from Voyager summed it up perfectly: **AI is about people first. Culture eats tech for breakfast.** The best tools in the world won't help unless you empower your team, model leadership, and champion a culture of curiosity, learning, and adaptability. **The Opportunity for NZ SMEs** Kiwi businesses have always punched above their weight. With AI, the opportunity is bigger than ever, unlocking innovation, efficiency, and customer engagement at a scale we've never seen. So get inspired, get practical, and ride the wave. **The leaders who invest now, building knowledge, capability, and culture will unlock extraordinary growth. AI is waiting for us to use it, understand it, and grow with it.** The Business North Harbour Conference proved it: **the time is now for New Zealand SMEs to become AI-powered and lead the world by example.** ## Power up your potential with practical AI skills. Contact us to discuss how Artificial Intelligence could boost your business. [Book an AI Chat](https://www.newzealand.ai/contact-us/?service=ai-chat#book) [Explore the AI Operating System](https://www.newzealand.ai/ai-os/) --- # AI in Aotearoa in 2025 - By the numbers Source: https://www.newzealand.ai/ai-in-aotearoa-in-2025-by-the-numbers/ 15 December 2025 · Insights # Government Strategy & Policy Reports **Justin Flitter** · Founder, AI New Zealand · 15 December 2025 · 9 min read We analysed the analysis, reviewed the research, scoured over the surveys to create a comprehensive round-up of the AI adoption reports published in 2025 about New Zealand. All up Perplexity references 145 sources from this deep-dive, all listed below. The picture they paint is genuinely exciting. Aotearoa's businesses are embracing AI at a cracking pace, though there's still work to be done on trust and skills. **AI adoption in New Zealand has reached a tipping point in 2025**. We've moved from early experimentation to mainstream integration, with 82-87% of businesses now using AI in some capacity. *// This is an extremely broad statement. LLM's integrated into your CRM or using ChatGPT as a Google Search replacement should not be classed as integration.* **The good news:** Productivity gains are real and measurable, 93% of businesses (surveyed) report efficiency improvements, and many are seeing substantial cost savings. The technology is becoming more accessible, with 70% keeping setup costs under $5,000. *// In one report only 40% of businesses reported an measurable impact at all from pilot projects. And their lies a hint. Those businesses really noticing the impact are moving past experimentation, not just embedding AI into daily workflows, but re-engineering workflows - occasionally eliminating tasks altogether.* **The challenges:** Trust remains fragile (only 34% of Kiwis trust AI), there's a significant SME adoption gap (68% have no plans to invest), and a critical skills shortage persists. Only 12% of organisations have successfully scaled AI across their entire operations. The momentum is undeniable, and 2025 marks the year NZ got serious about AI adoption. For businesses willing to invest with confidence, the productivity gains are there for the taking. [We're here to help with that...](https://academy.newzealand.ai/) ## **New Zealand's AI Strategy: Investing with Confidence** (July 2025) Published by the Ministry of Business, Innovation and Employment (MBIE), this is New Zealand's first-ever national AI strategy, making us the last OECD country to release one. [beehive](https://www.beehive.govt.nz/sites/default/files/2025-07/New%20Zealand's%20AI%20Strategy%20-%20Investing%20with%20confidence.pdf) **Key Findings:** - **67% of large NZ businesses** now use some form of AI, up dramatically from 48% in 2023 - However, **68% of SMEs have no plans** to evaluate or invest in AI, a significant adoption gap - Generative AI could add **$76 billion to the economy by 2038** (over 15% of GDP) - Only 34% of workers can clearly explain what AI is, despite 97% having heard of it - 43% of non-users cite lack of expertise as their main barrier **Strategic Focus:** The Government deliberately emphasises AI adoption over foundational development, positioning NZ as "smart adopters" who tailor global AI tools to local challenges in agriculture, healthcare, and education. **Budget Allocations:** - $213 million for tuition and training subsidies - $64 million for STEM and priority skills - $611 million in AI-related R&D projects approved since 2019 ## **2025 Cross-Agency AI Survey** (August 2025) Released by the Government Chief Digital Officer, this survey tracked AI use across the public sector. **Key Findings:** - **272 AI use cases** across 70 government agencies, more than doubling from 108 cases in 2024 - **55 AI cases now fully operational**, up from just 15 in 2024 - Increased productivity and efficiency reported across public service - Technological barriers remain a challenge ## Business & Industry Reports ## **AI in Action: New Zealand's Third AI Productivity Report** (August 2025) The AI Forum NZ, in partnership with Callaghan Innovation and Victoria University of Wellington, released their third bi-annual report. [aiforum](https://aiforum.org.nz/ais-impact-on-productivity-in-aotearoa/) **Key Findings:** - **91% of businesses report efficiency improvements** from AI - **77% of organisations report operational cost savings** - Over **25% achieved annual cost savings exceeding $50,000** - **70% have AI setup costs under $5,000**, making adoption accessible - **55% of businesses** say AI has created new career opportunities - **14% report job losses**, highlighting the need for upskilling - **93% report increased worker efficiency**, far exceeding global averages **Previous Reports in the Series:** - **First report (2024):** 67% adoption, almost all saw improved worker efficiency - **Second report (March 2025):** 15% surge in adoption over six months, only 7% job displacement ## **Datacom's State of AI Index 2025** (July 2025) This major survey of 200 senior leaders from large organisations and public sector bodies shows rapid acceleration. **Key Findings:** - **87% of organisations now use some form of AI**, up from 66% in 2024 and 48% in 2023 - **88% report positive impact** on operations [datacom](https://datacom.com/nz/en/solutions/artificial-intelligence/ai-insights/state-of-ai-index-2025) - **28% use advanced AI** - Only **12% have scaled AI across the entire organisation** - **46% are still in pilot phase**, 33% at department level **Maturity Gap:** While adoption is high, scaling remains a challenge, most firms are still exploring rather than fully embedding AI. ## **Kinetics: AI-Driven Productivity Gains in New Zealand** (June 2025) This detailed analysis examined AI's impact across diverse NZ industries. [kinetics](https://kinetics.co.nz/ai-driven-productivity-gains-in-new-zealand-2025/) **Key Statistics:** - **82% of NZ organisations** report using AI in operations - **93% say AI made workers more efficient** - **71% achieved operational cost savings** - Only **7% report AI directly replacing workers** - **29% set productivity as top priority for 2025** - **46% of firms** plan to boost AI investments, 41% focus on automation **Sector Highlights:** - Agriculture: AI expected to generate **$2.1 billion by 2035** through agritech applications - **Finance:** Fraud detection saving considerable analyst hours - **Manufacturing:** Predictive maintenance and quality control ## **Microsoft & Mandala: Unlocking New Zealand's $3.4 Billion AI Advantage** (June 2025) This partnership research unpacked how NZ can capitalise on AI opportunities. **Key Projections:** - AI applications and datacentre infrastructure could become a **$3.4 billion industry by 2035** - Generative AI adoption could increase **GDP per capita by 14%**, helping NZ leapfrog Finland and Estonia - **A public sector pilot of Microsoft Copilot showed 1781% ROI**, saving workers 13 hours monthly - Moderate automation scenario: **$76 billion annually by 2038** **Strategic Recommendations:** Focus on AI applications (healthtech, agritech, fintech) and infrastructure rather than competing on foundational models. ## Trust & Consumer Sentiment Reports ## **KPMG Trust in AI Survey Report 2025** (University of Melbourne & KPMG) This global study surveyed 48,000+ people across 47 countries, including New Zealand. **New Zealand Findings:** - Only **34% of Kiwis trust AI** systems [kpmg](https://assets.kpmg.com/content/dam/kpmgsites/xx/pdf/2025/05/trust-attitudes-and-use-of-ai-new-zealand-snapshot.pdf.coredownload.inline.pdf) - **48% accept or approve of AI** - **81% believe AI regulation is needed** - **85% more willing to trust AI** when there's clear accountability - **43% report increased efficiency** from AI at work - **31% report increased revenue-generating activity** **Trust Gap:** Many NZ workers use AI regularly but don't trust it, creating complex organisational risks. ## **One NZ AI Trust Report 2025** (May 2025) Survey of 1,000 New Zealanders examining trust and adoption. **Key Findings:** - **77% have knowingly used AI-powered services** in the past year - Nearly **half don't trust large companies** to use AI ethically. [one](https://media.one.nz/aireport) - **62% would stop using a company** if concerned about its AI use - **87% believe strong data privacy protections** are important or essential **Top Concerns:** - 67% concerned about misuse of personal data - 65% worried about job losses - 62% concerned about unfair decision-making - 48% anxious about AI energy consumption and greenwashing **Opportunities Seen:** - 43% see value in AI for boosting productivity - 42% for cybersecurity - 41% for innovation **Sector Trust Challenges:** Government, banking, and healthcare face greater challenges building public confidence in AI use. ## **Ipsos AI Monitor 2025** (October 2025) This 31-country study of 24,000+ people included New Zealand. [ipsos](https://www.ipsos.com/en-nz/understanding-aotearoa-new-zealand-ipsos-ai-monitor-2025) **Key Findings:** - **73% of New Zealanders report good understanding of AI**, up from 69% in 2024 and 62% in 2023 - **66% say AI makes them nervous** - second highest globally after Australia at 67% - Understanding is rising but so is nervousness ## **EY AI Sentiment Report for New Zealand and Australia** (January 2025) This report explored perceptions across both countries. [ey](https://www.ey.com/content/dam/ey-unified-site/ey-com/aus-nzl/documents/pdfs/ey-ai-sentiment-index-nz-and-au-2025.pdf) **Key Insights:** - Trust deficit and blurring of reality are major concerns - Baby Boomers and Gen X more sceptical of AI than younger generations - Sectors like education and healthcare set to be revolutionised, but trust remains crucial ## Sector-Specific Reports ## **World Internet Project New Zealand (WIP-NZ 2025)** (Published by AUT) This longitudinal survey examined internet connectivity, AI adoption, and digital attitudes. **Focus Areas:** - Quality and accessibility of internet connections in NZ - Adoption and trust in AI systems - New Zealanders' attitudes towards digital technologies - Daily social media use dropped from 83% (2023) to 72% (2025) [istart](https://istart.co.nz/nz-news-items/ai-takes-centre-stage-while-websites-lag/) ## **AI in Primary Care Survey – Round Two** (February 2025) Joint survey by the AI in Primary Care group examining healthcare adoption. **Key Findings:** - **68% of primary care respondents now use AI tools**, up from 52% in Round One. [gpnz](https://gpnz.org.nz/media-releases/joint-media-release-round-two-ai-in-primary-care-survey-highlights-growing-adoption/) - **Daily AI use rose from 25% to 40%** - Growing contribution of AI in reducing administrative burden - Barriers remain for widespread adoption ## **InternetNZ: NZ Consumers and Businesses Research** (August 2025) Survey examining online presence and AI integration. [internetnz](https://internetnz.nz/assets/DotNZ-Research-2025-Summary.pdf) **Business Findings:** - **68% of businesses are using AI** - AI integration focused on time-saving tools, customer support automation, and marketing personalisation - Businesses notice consumers demanding faster, more personalised interactions and increased self-service tools ## Professional Services Firm Reports ## **PwC CEO Survey 2025** (March 2025) Survey of New Zealand CEOs examining business sentiment and AI adoption. **Key Findings:** - **70% of NZ CEOs report AI enhances employee work efficiency**, significantly higher than Australia (42%), Asia Pacific (58%), and globally (56%) - **55% report AI increases efficiency in their own work time** [pwc](https://www.pwc.co.nz/insights-and-publications/ceo-survey/ceo-survey-2025.html) - **47% believe AI will boost company profitability** over next 12 months - Only **13% consider AI "extremely important"** for long-term success, suggesting cautious approach - **28% express high degree of trust** in integrating AI into business practices **Strategic Priorities:** CEOs prioritising integration of AI into technology platforms, business processes, and workflows over next three years. ## PwC Global AI Jobs Barometer 2025 – New Zealand Analysis Examined how AI is changing job skills and employment. **Key Finding:** - Skills sought by employers for AI-exposed jobs are changing **66% faster** than for other jobs, up from 25% last year [pwc](https://www.pwc.com/gx/en/issues/artificial-intelligence/job-barometer/aijb-2025-new-zealand-analysis.pdf) - AI associated with gentler growth patterns ## **PwC Insurance Banana Skins 2025** (November 2025) Survey of 21 insurance companies in New Zealand. **Key Finding:** - **AI jumped from #10 risk (2023) to #1 risk (2025)** for NZ insurers [pwc](https://www.pwc.co.nz/insights-and-publications/2025-publications/new-zealand-insurance-banana-skins-2025.html) - Cybercrime ranked #2 locally (#1 globally) - Climate change ranked #3 locally **Recommendation:** Focus on upskilling people to better understand AI and develop governance frameworks. ## **KPMG CEO Outlook 2025** (November 2025) Survey of global and NZ CEOs **New Zealand Findings:** - **90% of NZ CEOs** feel their leadership has clear view on how AI will disrupt business models - higher than global average [kpmg](https://kpmg.com/nz/en/insights/leadership/ceo-outlook-2025.html) - Clear agreement that upskilling and regulatory progress are key to success ## **Deloitte New Zealand Impact Report 2025** (November 2025) Overview of Deloitte's AI work and broader business impact **Focus:** Building trusted intelligent government, AI governance, and business transformation through AI. ## Academic & Research Reports ## **Digital Revolution in Business Study** (May 2025) Research examining AI adoption's impact on business performance and innovation in NZ. **Key Findings:** - Digital infrastructure has strong positive association with business performance and innovation capabilities. [brdr](https://brdr.org/index.php/researchrise/article/view/113) - AI linked to driving business through automation, predictive analytics, and strategic decision-making - Government digital policies imperative in creating favourable environment for digital transformation ## **Smart Solutions for Tough Times: AI in NZ's Food & Beverage Industry** (September 2025) Published by AUT's Hospitality Insights, this study explored AI opportunities and challenges in F&B sector. **Focus Areas:** - Inventory management through predictive analytics - Operational efficiency in food production - Customer service enhancement - Revenue maximisation ## **Geospatial Technologies in NZ's Plantation Forestry Sector** (November 2025) Third survey following 2013 and 2018 studies. **Key Findings:** - **27 of 29 organisations responded** (93% response rate) - Organisations manage 1,283,000 hectares (74% of NZ's plantation forest estate) - 100% use aerial imagery from online portals [nzjforestryscience](https://nzjforestryscience.nz/index.php/nzjfs/article/view/430) - AI and remote sensing technologies increasingly adopted - [https://www.beehive.govt.nz/sites/default/files/2025-07/New%20Zealand's%20AI%20Strategy%20-%20Investing%20with%20confidence.pdf](https://www.beehive.govt.nz/sites/default/files/2025-07/New%20Zealand's%20AI%20Strategy%20-%20Investing%20with%20confidence.pdf) - [https://news.microsoft.com/source/asia/features/nz-new-ai-economy-report-2025/](https://news.microsoft.com/source/asia/features/nz-new-ai-economy-report-2025/) - [https://www.digital.govt.nz/standards-and-guidance/technology-and-architecture/artificial-intelligence/research-ai-public-service/2025-cross-agency-ai-survey-highlights](https://www.digital.govt.nz/standards-and-guidance/technology-and-architecture/artificial-intelligence/research-ai-public-service/2025-cross-agency-ai-survey-highlights) - [https://www.newzealand.ai/new-zealand-ai-adoption-in-2025](https://www.newzealand.ai/new-zealand-ai-adoption-in-2025/) - [https://www.mbie.govt.nz/business-and-employment/economic-growth/digital-policy/new-zealands-ai-strategy-investing-with-confidence/addressing-barriers-to-ai-uptake-in-new-zealand](https://www.mbie.govt.nz/business-and-employment/economic-growth/digital-policy/new-zealands-ai-strategy-investing-with-confidence/addressing-barriers-to-ai-uptake-in-new-zealand) - [https://aiforum.org.nz/ais-impact-on-productivity-in-aotearoa/](https://aiforum.org.nz/ais-impact-on-productivity-in-aotearoa/) - [https://aiforum.org.nz/reports/ai-in-action-key-findings-from-new-zealands-third-ai-productivity-report/](https://aiforum.org.nz/reports/ai-in-action-key-findings-from-new-zealands-third-ai-productivity-report/) - [https://www.linkedin.com/posts/mahsamohaghegh_ai-in-action-report-3-activity-7365947795859505152-eEme](https://www.linkedin.com/posts/mahsamohaghegh_ai-in-action-report-3-activity-7365947795859505152-eEme) - [https://kinetics.co.nz/ai-driven-productivity-gains-in-new-zealand-2025/](https://kinetics.co.nz/ai-driven-productivity-gains-in-new-zealand-2025/) - [https://aigovernance.aiforum.org.nz/ais-impact-on-productivity-in-aotearoa/](https://aigovernance.aiforum.org.nz/ais-impact-on-productivity-in-aotearoa/) - [https://thewebcompany.digital/ai-new-zealand-2025-state-adoption-barriers/](https://thewebcompany.digital/ai-new-zealand-2025-state-adoption-barriers/) - [https://itbrief.co.nz/story/ai-adoption-boosts-productivity-across-new-zealand-businesses](https://itbrief.co.nz/story/ai-adoption-boosts-productivity-across-new-zealand-businesses) - [https://aotearoaai.nz/ai-adoption-in-aotearoa-is-accelerating-now-its-time-to-scale/](https://aotearoaai.nz/ai-adoption-in-aotearoa-is-accelerating-now-its-time-to-scale/) - [https://datacom.com/nz/en/solutions/artificial-intelligence/ai-insights/state-of-ai-index-2025](https://datacom.com/nz/en/solutions/artificial-intelligence/ai-insights/state-of-ai-index-2025) - [https://itbrief.co.nz/story/new-zealand-firms-see-ai-boost-but-scaling-hurdles-remain](https://itbrief.co.nz/story/new-zealand-firms-see-ai-boost-but-scaling-hurdles-remain) - [https://digital.nemko.com/news/new-zealand-ai-strategy-2025](https://digital.nemko.com/news/new-zealand-ai-strategy-2025) - [https://www.reseller.co.nz/article/4011310/microsoft-study-worker-reallocation-one-key-to-unlocking-ai-benefits-in-new-zealand.html](https://www.reseller.co.nz/article/4011310/microsoft-study-worker-reallocation-one-key-to-unlocking-ai-benefits-in-new-zealand.html) - [https://kpmg.com/nz/en/insights/ai/trust-attitudes-and-use-of-ai.html](https://kpmg.com/nz/en/insights/ai/trust-attitudes-and-use-of-ai.html) - [https://www.reseller.co.nz/article/3977703/kpmg-finds-kiwis-embrace-ai-at-work-but-trust-and-governance-pose-challenges.html](https://www.reseller.co.nz/article/3977703/kpmg-finds-kiwis-embrace-ai-at-work-but-trust-and-governance-pose-challenges.html) - [https://assets.kpmg.com/content/dam/kpmgsites/xx/pdf/2025/05/trust-attitudes-and-use-of-ai-new-zealand-snapshot.pdf.coredownload.inline.pdf](https://assets.kpmg.com/content/dam/kpmgsites/xx/pdf/2025/05/trust-attitudes-and-use-of-ai-new-zealand-snapshot.pdf.coredownload.inline.pdf) - [https://media.one.nz/aireport](https://media.one.nz/aireport) - [https://telconews.co.nz/story/new-zealanders-demand-trust-ethics-in-ai-adoption-study-finds](https://telconews.co.nz/story/new-zealanders-demand-trust-ethics-in-ai-adoption-study-finds) - [https://www.reseller.co.nz/article/3995181/one-nz-research-reveals-unease-about-ai-impacts.html](https://www.reseller.co.nz/article/3995181/one-nz-research-reveals-unease-about-ai-impacts.html) - [https://www.newstalkzb.co.nz/on-air/heather-du-plessis-allan-drive/audio/jason-paris-one-nz-chief-executive-on-the-new-report-revealing-many-kiwis-dont-trust-ai/](https://www.newstalkzb.co.nz/on-air/heather-du-plessis-allan-drive/audio/jason-paris-one-nz-chief-executive-on-the-new-report-revealing-many-kiwis-dont-trust-ai/) - [https://www.ipsos.com/en-nz/understanding-aotearoa-new-zealand-ipsos-ai-monitor-2025](https://www.ipsos.com/en-nz/understanding-aotearoa-new-zealand-ipsos-ai-monitor-2025) - [https://www.ey.com/content/dam/ey-unified-site/ey-com/aus-nzl/documents/pdfs/ey-ai-sentiment-index-nz-and-au-2025.pdf](https://www.ey.com/content/dam/ey-unified-site/ey-com/aus-nzl/documents/pdfs/ey-ai-sentiment-index-nz-and-au-2025.pdf) - [https://www.ey.com/en_nz/insights/ai/ai-in-the-workforce-bridging-technology-and-human-elements](https://www.ey.com/en_nz/insights/ai/ai-in-the-workforce-bridging-technology-and-human-elements) - [https://nzpri.aut.ac.nz/__data/assets/pdf_file/0008/1073906/WIP-NZ-Report-2025_InternetNZ.pdf](https://nzpri.aut.ac.nz/__data/assets/pdf_file/0008/1073906/WIP-NZ-Report-2025_InternetNZ.pdf) - [https://www.digitalequity.nz/resources-content/the-world-internet-project-in-new-zealand-aut](https://www.digitalequity.nz/resources-content/the-world-internet-project-in-new-zealand-aut) - [https://istart.co.nz/nz-news-items/ai-takes-centre-stage-while-websites-lag/](https://istart.co.nz/nz-news-items/ai-takes-centre-stage-while-websites-lag/) - [https://gpnz.org.nz/media-releases/joint-media-release-round-two-ai-in-primary-care-survey-highlights-growing-adoption/](https://gpnz.org.nz/media-releases/joint-media-release-round-two-ai-in-primary-care-survey-highlights-growing-adoption/) - [https://internetnz.nz/assets/DotNZ-Research-2025-Summary.pdf](https://internetnz.nz/assets/DotNZ-Research-2025-Summary.pdf) - [https://www.pwc.co.nz/insights-and-publications/ceo-survey/ceo-survey-2025.html](https://www.pwc.co.nz/insights-and-publications/ceo-survey/ceo-survey-2025.html) - [https://www.hcamag.com/nz/specialisation/benefits/seven-in-10-ceos-said-ai-boosted-employees-time-at-work-report/536226](https://www.hcamag.com/nz/specialisation/benefits/seven-in-10-ceos-said-ai-boosted-employees-time-at-work-report/536226) - [https://www.pwc.com/gx/en/issues/artificial-intelligence/job-barometer/aijb-2025-new-zealand-analysis.pdf](https://www.pwc.com/gx/en/issues/artificial-intelligence/job-barometer/aijb-2025-new-zealand-analysis.pdf) - [https://www.reseller.co.nz/article/4088898/ai-jumps-to-top-concern-for-new-zealand-insurers-report.html](https://www.reseller.co.nz/article/4088898/ai-jumps-to-top-concern-for-new-zealand-insurers-report.html) - [https://www.pwc.co.nz/insights-and-publications/2025-publications/new-zealand-insurance-banana-skins-2025.html](https://www.pwc.co.nz/insights-and-publications/2025-publications/new-zealand-insurance-banana-skins-2025.html) - [https://kpmg.com/nz/en/insights/leadership/ceo-outlook-2025.html](https://kpmg.com/nz/en/insights/leadership/ceo-outlook-2025.html) - [https://www.deloitte.com/nz/en/about/governance/new-zealand-impact-report.html](https://www.deloitte.com/nz/en/about/governance/new-zealand-impact-report.html) - [https://www.deloitte.com/nz/en/about/governance/new-zealand-impact-report/business.html](https://www.deloitte.com/nz/en/about/governance/new-zealand-impact-report/business.html) - [https://www.deloitte.com/nz/en/services/consulting/collections/deloitte-ai-institute.html](https://www.deloitte.com/nz/en/services/consulting/collections/deloitte-ai-institute.html) - [https://www.deloitte.com/nz/en/Industries/consumer/analysis/trustworthy-artificial-intelligence.html](https://www.deloitte.com/nz/en/Industries/consumer/analysis/trustworthy-artificial-intelligence.html) - [https://brdr.org/index.php/researchrise/article/view/113](https://brdr.org/index.php/researchrise/article/view/113) - [https://ojs.aut.ac.nz/hospitality-insights/article/view/177](https://ojs.aut.ac.nz/hospitality-insights/article/view/177) - [https://nzjforestryscience.nz/index.php/nzjfs/article/view/430](https://nzjforestryscience.nz/index.php/nzjfs/article/view/430) - [https://periodicos.ufrn.br/ambiente/article/view/38676](https://periodicos.ufrn.br/ambiente/article/view/38676) - [https://neurologyopen.bmj.com/lookup/doi/10.1136/bmjno-2025-001361](https://neurologyopen.bmj.com/lookup/doi/10.1136/bmjno-2025-001361) - [https://onlinelibrary.wiley.com/doi/10.1002/ece3.72510](https://onlinelibrary.wiley.com/doi/10.1002/ece3.72510) - [https://www.ewadirect.com/proceedings/aemps/article/view/21166](https://www.ewadirect.com/proceedings/aemps/article/view/21166) - [https://link.springer.com/10.1007/s11192-025-05468-y](https://link.springer.com/10.1007/s11192-025-05468-y) - [https://www.unitec.ac.nz/epress/wp-content/uploads/2025/11/LP_AIoT_250115.pdf](https://www.unitec.ac.nz/epress/wp-content/uploads/2025/11/LP_AIoT_250115.pdf) - [https://www.tandfonline.com/doi/full/10.1080/03036758.2025.2455499](https://www.tandfonline.com/doi/full/10.1080/03036758.2025.2455499) - [https://arxiv.org/abs/2504.07139](https://arxiv.org/abs/2504.07139) - [https://arxiv.org/pdf/2311.15473.pdf](https://arxiv.org/pdf/2311.15473.pdf) - [https://linkinghub.elsevier.com/retrieve/pii/S0268401222001220](https://linkinghub.elsevier.com/retrieve/pii/S0268401222001220) - [https://arxiv.org/pdf/1504.07012.pdf](https://arxiv.org/pdf/1504.07012.pdf) - [https://arxiv.org/pdf/2503.09613.pdf](https://arxiv.org/pdf/2503.09613.pdf) - [https://www.tandfonline.com/doi/pdf/10.1080/08839514.2024.2318670?needAccess=true](https://www.tandfonline.com/doi/pdf/10.1080/08839514.2024.2318670?needAccess=true) - [http://arxiv.org/pdf/2302.14139.pdf](http://arxiv.org/pdf/2302.14139.pdf) - [https://pmc.ncbi.nlm.nih.gov/articles/PMC10474148/](https://pmc.ncbi.nlm.nih.gov/articles/PMC10474148/) - [https://www.digital.govt.nz/standards-and-guidance/technology-and-architecture/artificial-intelligence/new-zealands-ai-strategy-and-guidance-for-business](https://www.digital.govt.nz/standards-and-guidance/technology-and-architecture/artificial-intelligence/new-zealands-ai-strategy-and-guidance-for-business) - [https://www.mbie.govt.nz/business-and-employment/economic-growth/digital-policy/new-zealands-ai-strategy-investing-with-confidence](https://www.mbie.govt.nz/business-and-employment/economic-growth/digital-policy/new-zealands-ai-strategy-investing-with-confidence) - [https://openaccess.wgtn.ac.nz/articles/report/AI_in_Action_Exploring_the_Impact_of_Artificial_Intelligence_on_New_Zealand_s_Productivity_Report_2_/29916128/1/files/57198197.pdf](https://openaccess.wgtn.ac.nz/articles/report/AI_in_Action_Exploring_the_Impact_of_Artificial_Intelligence_on_New_Zealand_s_Productivity_Report_2_/29916128/1/files/57198197.pdf) - [https://aiforum.org.nz/wp-content/uploads/2025/08/AI-Forum-Productivity-Report_Website_Aug-2025.pdf](https://aiforum.org.nz/wp-content/uploads/2025/08/AI-Forum-Productivity-Report_Website_Aug-2025.pdf) - [https://www.myob.com/nz/insight/post/midsized-business-ai-payoff](https://www.myob.com/nz/insight/post/midsized-business-ai-payoff) - [https://itbrief.co.nz/story/ai-adoption-helps-nz-smes-boost-sales-cut-marketing-costs](https://itbrief.co.nz/story/ai-adoption-helps-nz-smes-boost-sales-cut-marketing-costs) - [https://www.semanticscholar.org/paper/3cd1e0010029d73ca52f0b8a4e81dd15d24d3b9f](https://www.semanticscholar.org/paper/3cd1e0010029d73ca52f0b8a4e81dd15d24d3b9f) - [https://hdl.handle.net/10289/17278](https://hdl.handle.net/10289/17278) - [https://journalajeba.com/index.php/AJEBA/article/view/1919](https://journalajeba.com/index.php/AJEBA/article/view/1919) - [https://www.nursingpraxis.org/article/140633-new-zealand-nurses-use-and-perception-of-digital-health-during-the-covid-19-pandemic](https://www.nursingpraxis.org/article/140633-new-zealand-nurses-use-and-perception-of-digital-health-during-the-covid-19-pandemic) - [https://www.cureus.com/articles/420048-a-pilot-study-of-british-and-irish-junior-doctors-in-auckland-new-zealand-using-a-questionnaire-to-explore-motivations-and-intentions](https://www.cureus.com/articles/420048-a-pilot-study-of-british-and-irish-junior-doctors-in-auckland-new-zealand-using-a-questionnaire-to-explore-motivations-and-intentions) - [https://www.nursingpraxis.org/article/144221-evaluation-of-the-aotearoa-new-zealand-transition-programme-for-nurse-practitioners-in-primary-healthcare](https://www.nursingpraxis.org/article/144221-evaluation-of-the-aotearoa-new-zealand-transition-programme-for-nurse-practitioners-in-primary-healthcare) - [https://repositorio.bde.es/handle/123456789/40126](https://repositorio.bde.es/handle/123456789/40126) - [https://www.frontiersin.org/articles/10.3389/frai.2025.1606431/full](https://www.frontiersin.org/articles/10.3389/frai.2025.1606431/full) - [https://linkinghub.elsevier.com/retrieve/pii/S0148296324000468](https://linkinghub.elsevier.com/retrieve/pii/S0148296324000468) - [https://www.mdpi.com/2071-1050/13/4/2025/pdf?version=1613977418](https://www.mdpi.com/2071-1050/13/4/2025/pdf?version=1613977418) - [http://anapub.co.ke/journals/jebi/jebi_pdf/2023/jebi_volume_3-issue_1/JEBI202303005.pdf](http://anapub.co.ke/journals/jebi/jebi_pdf/2023/jebi_volume_3-issue_1/JEBI202303005.pdf) - [https://ijsra.net/sites/default/files/IJSRA-2024-0235.pdf](https://ijsra.net/sites/default/files/IJSRA-2024-0235.pdf) - [http://arxiv.org/pdf/2401.02843.pdf](http://arxiv.org/pdf/2401.02843.pdf) - [https://www.mdpi.com/2673-2688/1/2/11/pdf](https://www.mdpi.com/2673-2688/1/2/11/pdf) - [https://www.mbie.govt.nz/science-and-technology/science-and-innovation/refocusing-the-science-innovation-and-technology-system/public-research-organisations/new-zealand-institute-for-advanced-technology/call-for-proposals-for-the-artificial-intelligence-research-platform](https://www.mbie.govt.nz/science-and-technology/science-and-innovation/refocusing-the-science-innovation-and-technology-system/public-research-organisations/new-zealand-institute-for-advanced-technology/call-for-proposals-for-the-artificial-intelligence-research-platform) - [https://nztech.org.nz/2025/09/01/ai-in-action-how-nz-is-gaining-real-wins/](https://nztech.org.nz/2025/09/01/ai-in-action-how-nz-is-gaining-real-wins/) - [https://ojs.victoria.ac.nz/pq/article/view/9933](https://ojs.victoria.ac.nz/pq/article/view/9933) - [https://arxiv.org/abs/2510.13653](https://arxiv.org/abs/2510.13653) - [https://www.semanticscholar.org/paper/1cd33284040a84acef58abd227ecca19bdf76c1d](https://www.semanticscholar.org/paper/1cd33284040a84acef58abd227ecca19bdf76c1d) - [https://arxiv.org/pdf/2310.02521.pdf](https://arxiv.org/pdf/2310.02521.pdf) - [https://www.mdpi.com/2071-1050/16/5/1790/pdf?version=1708581624](https://www.mdpi.com/2071-1050/16/5/1790/pdf?version=1708581624) - [https://arxiv.org/pdf/2308.16538.pdf](https://arxiv.org/pdf/2308.16538.pdf) - [https://arxiv.org/pdf/2403.17405.pdf](https://arxiv.org/pdf/2403.17405.pdf) - [https://journals.sagepub.com/doi/pdf/10.1177/00081256231190430](https://journals.sagepub.com/doi/pdf/10.1177/00081256231190430) - [https://kpmg.com/nz/en/insights/ai/new-zealand--are-you-ai-ready--.html](https://kpmg.com/nz/en/insights/ai/new-zealand--are-you-ai-ready--.html) - [https://news.microsoft.com/source/asia/2025/12/03/ai-can-help-rewrite-new-zealands-productivity-story-with-the-right-prompts/](https://news.microsoft.com/source/asia/2025/12/03/ai-can-help-rewrite-new-zealands-productivity-story-with-the-right-prompts/) - [https://www.linkedin.com/posts/deloitte_deloittenz-activity-7403901189811032064-v4Nq](https://www.linkedin.com/posts/deloitte_deloittenz-activity-7403901189811032064-v4Nq) - [https://kpmg.com/nz/en/insights/government-public-sector/building-a-trusted-intelligent-government-in-new-zealand.html](https://kpmg.com/nz/en/insights/government-public-sector/building-a-trusted-intelligent-government-in-new-zealand.html) - [https://news.microsoft.com/source/asia/2025/10/21/digital-government-report-nz/](https://news.microsoft.com/source/asia/2025/10/21/digital-government-report-nz/) - [https://s-rsa.com/index.php/agi/article/view/16671](https://s-rsa.com/index.php/agi/article/view/16671) - [https://www.semanticscholar.org/paper/07132de299cb9d25d7b13be5828de935992a2591](https://www.semanticscholar.org/paper/07132de299cb9d25d7b13be5828de935992a2591) - [https://arxiv.org/pdf/2503.08725.pdf](https://arxiv.org/pdf/2503.08725.pdf) - [https://www.tandfonline.com/doi/pdf/10.1080/02642069.2024.2336208?needAccess=true](https://www.tandfonline.com/doi/pdf/10.1080/02642069.2024.2336208?needAccess=true) - [https://fyi.org.nz/request/31575/response/129739/attach/5/DD1%20National%20AI%20Strategy%20April%202025.pdf](https://fyi.org.nz/request/31575/response/129739/attach/5/DD1%20National%20AI%20Strategy%20April%202025.pdf) - [https://www.spark.co.nz/online/large-business-govt/why-choose-spark/why-spark/the-state-of-cloud-2025](https://www.spark.co.nz/online/large-business-govt/why-choose-spark/why-spark/the-state-of-cloud-2025) - [https://www.pwc.co.nz/insights-and-publications/2025-publications/audit-transparency-report-2025.html](https://www.pwc.co.nz/insights-and-publications/2025-publications/audit-transparency-report-2025.html) - [https://www.ey.com/en_nz/services/ai](https://www.ey.com/en_nz/services/ai) - [https://www.spark.co.nz/content/dam/spark/documents/pdfs/accelerate-aotearoa/lifting-productivity-report.pdf](https://www.spark.co.nz/content/dam/spark/documents/pdfs/accelerate-aotearoa/lifting-productivity-report.pdf) - [https://www.spark.co.nz/online/large-business-govt/why-choose-spark/insights/cloud-as-a-catalyst](https://www.spark.co.nz/online/large-business-govt/why-choose-spark/insights/cloud-as-a-catalyst) - [https://www.ey.com/en_nz/insights/workforce/work-reimagined-survey](https://www.ey.com/en_nz/insights/workforce/work-reimagined-survey) - [https://www.spark.co.nz/online/about/our-company/news/Cloud_spend_to_hit_9B_by_2028](https://www.spark.co.nz/online/about/our-company/news/Cloud_spend_to_hit_9B_by_2028) - [https://www.pwc.co.nz/insights-and-publications/2025-publications/new-zealand-race-for-growth-ai-climate-global-shifts.html](https://www.pwc.co.nz/insights-and-publications/2025-publications/new-zealand-race-for-growth-ai-climate-global-shifts.html) - [https://www.ey.com/en_nz/insights/tech-sector/top-10-opportunities-for-technology-companies-in-2025](https://www.ey.com/en_nz/insights/tech-sector/top-10-opportunities-for-technology-companies-in-2025) - [https://investors.sparknz.co.nz/FormBuilder/_Resource/_module/gXbeer80tkeL4nEaF-kwFA/doc/FY25_Annual_Report.pdf](https://investors.sparknz.co.nz/FormBuilder/_Resource/_module/gXbeer80tkeL4nEaF-kwFA/doc/FY25_Annual_Report.pdf) - [https://www.linkedin.com/posts/matthew-cowie-285894_new-zealands-opportunity-to-lead-in-clean-activity-7322553520686014464-UR5d](https://www.linkedin.com/posts/matthew-cowie-285894_new-zealands-opportunity-to-lead-in-clean-activity-7322553520686014464-UR5d) - [https://www.semanticscholar.org/paper/29c1bf3757c52be4493b4d75fef2d3e7b255c90b](https://www.semanticscholar.org/paper/29c1bf3757c52be4493b4d75fef2d3e7b255c90b) - [https://ijlter.org/index.php/ijlter/article/view/15048](https://ijlter.org/index.php/ijlter/article/view/15048) - [https://link.springer.com/10.1007/s00784-025-06618-5](https://link.springer.com/10.1007/s00784-025-06618-5) - [https://link.springer.com/10.1007/s11192-025-05446-4](https://link.springer.com/10.1007/s11192-025-05446-4) - [https://aacrjournals.org/cancerres/article/85/8_Supplement_1/1049/756648/Abstract-1049-Advancing-health-equity-in-oncology](https://aacrjournals.org/cancerres/article/85/8_Supplement_1/1049/756648/Abstract-1049-Advancing-health-equity-in-oncology) - [https://invergejournals.com/index.php/ijss/article/view/200](https://invergejournals.com/index.php/ijss/article/view/200) - [https://www.nature.com/articles/s41433-025-03850-x](https://www.nature.com/articles/s41433-025-03850-x) - [https://www.researchprotocols.org/2025/1/e80290](https://www.researchprotocols.org/2025/1/e80290) - [https://pmc.ncbi.nlm.nih.gov/articles/PMC8480835/](https://pmc.ncbi.nlm.nih.gov/articles/PMC8480835/) - [https://anzswjournal.nz/anzsw/article/download/481/597](https://anzswjournal.nz/anzsw/article/download/481/597) - [https://arxiv.org/ftp/arxiv/papers/2103/2103.06312.pdf](https://arxiv.org/ftp/arxiv/papers/2103/2103.06312.pdf) - [https://pmc.ncbi.nlm.nih.gov/articles/PMC8362386/](https://pmc.ncbi.nlm.nih.gov/articles/PMC8362386/) - [https://arxiv.org/abs/2405.19522](https://arxiv.org/abs/2405.19522) - [https://arxiv.org/pdf/2210.15767.pdf](https://arxiv.org/pdf/2210.15767.pdf) - [https://www.rnz.co.nz/news/national/574517/universities-give-up-using-software-to-detect-ai-in-students-work](https://www.rnz.co.nz/news/national/574517/universities-give-up-using-software-to-detect-ai-in-students-work) - [https://www.aut.ac.nz/about/ai-at-aut/ai-research](https://www.aut.ac.nz/about/ai-at-aut/ai-research) - [https://www.auckland.ac.nz/en/news/2025/08/18/new-zealand-s-ai-advantage.html](https://www.auckland.ac.nz/en/news/2025/08/18/new-zealand-s-ai-advantage.html) - [https://www.aut.ac.nz/research/academic-departments/data-science-and-artificial-intelligence-department](https://www.aut.ac.nz/research/academic-departments/data-science-and-artificial-intelligence-department) - [https://www.aut.ac.nz/about/ai-at-aut](https://www.aut.ac.nz/about/ai-at-aut) - [https://aiforum.org.nz/event/aotearoa-ai-awards-2025/](https://aiforum.org.nz/event/aotearoa-ai-awards-2025/) - [https://research-hub.auckland.ac.nz/subhub/ai-in-research](https://research-hub.auckland.ac.nz/subhub/ai-in-research) - [https://www.auckland.ac.nz/en/news/2025/10/09/behind-the-big-promises-on-ai-in-business-lies----nothing-.html](https://www.auckland.ac.nz/en/news/2025/10/09/behind-the-big-promises-on-ai-in-business-lies----nothing-.html) - [https://www.aut.ac.nz/news/stories/trust-in-news-report-2025](https://www.aut.ac.nz/news/stories/trust-in-news-report-2025) - [https://www.auckland.ac.nz/en/news/2025/08/21/exploring-ai-s-environmental-costs-and-benefits.html](https://www.auckland.ac.nz/en/news/2025/08/21/exploring-ai-s-environmental-costs-and-benefits.html) - [https://nzpri.aut.ac.nz/research/primary-surveys/world-internet-project-nz](https://nzpri.aut.ac.nz/research/primary-surveys/world-internet-project-nz) - [https://ml4sg.auckland.ac.nz/reports/](https://ml4sg.auckland.ac.nz/reports/) - [https://arxiv.org/pdf/2304.11771.pdf](https://arxiv.org/pdf/2304.11771.pdf) - [https://arxiv.org/pdf/2309.11690.pdf](https://arxiv.org/pdf/2309.11690.pdf) - [https://arxiv.org/html/2503.04761v1](https://arxiv.org/html/2503.04761v1) - [https://www.tandfonline.com/doi/pdf/10.1080/09537287.2023.2234882?needAccess=true&role=button](https://www.tandfonline.com/doi/pdf/10.1080/09537287.2023.2234882?needAccess=true&role=button) - [https://arxiv.org/pdf/2410.18357.pdf](https://arxiv.org/pdf/2410.18357.pdf) - [https://www.linkedin.com/posts/onenz_ai-sustainability-trustintech-activity-7365661289592848385-X9LF](https://www.linkedin.com/posts/onenz_ai-sustainability-trustintech-activity-7365661289592848385-X9LF) - [https://media.one.nz](https://media.one.nz/) - [https://aiforum.org.nz/our-work/publications/](https://aiforum.org.nz/our-work/publications/) ## Power up your potential with practical AI skills. Contact us to discuss how Artificial Intelligence could boost your business. [Book an AI Chat](https://www.newzealand.ai/contact-us/?service=ai-chat#book) [Explore the AI Operating System](https://www.newzealand.ai/ai-os/) --- # The new AI platform features changing how we work Source: https://www.newzealand.ai/ai-platforms-are-releasing-new-functionality-that-s/ 15 March 2026 · Guidelines # The Four Platforms You Need to Know AI platforms are releasing new functionality that's changing the way we work. **Justin Flitter** · Founder, AI New Zealand · 15 March 2026 · 8 min read ## This is not an incremental improvement - it's a seismic leap. Here's what your business needs to understand right now. Twelve weeks ago, the best-practice approach to using AI in your business was: open a chat window, type a prompt, upload your document, review the output. That approach still works. But it's now the manual labourer's method, the equivalent of faxing when your competitors are running automated pipelines. In the first quarter of 2026, four major platforms crossed a threshold that most New Zealand leaders haven't yet registered. They stopped being AI assistants you talk to. They became AI workers you delegate to. ***The shift isn't a feature update. It's a change in the fundamental operating logic of how AI sits inside a business.*** The businesses racing ahead aren't using better prompts. They are enabling their people to understand context protocols, build persistent memory, and hand off multi-step work to AI agents that run while they sleep. This briefing covers four platforms that define what 'capable AI' looks like today. **What Changed in 90 Days** The shift can be summarised in three moves: The platforms below are the ones setting the new standard. Each one matters. Together, they define the gap that is opening right now between AI-theatre businesses and AI-operational businesses. ## **Claude Cowork** *- Agentic AI for knowledge workers* **What it is** Claude Cowork is Anthropic's answer to one question: what happens when you take the most powerful coding agent in the world (Claude Code, used by 115,000 developers) and give it to everyone else? Launched in January 2026 and built in approximately ten days, using Claude Code itself, Cowork brings autonomous, multi-step AI execution to non-technical knowledge workers. Instead of a chat, you get a task queue. Instead of a response, you get finished deliverables, formatted documents, organised files, synthesised research, Excel spreadsheets with working formulas, PowerPoint presentations. **What it does** - Reads from and writes to your local files without manual uploads or downloads - Breaks complex work into subtasks and coordinates parallel sub-agents simultaneously - Runs long-duration tasks without timeouts, you describe the outcome, step away, and return to finished work - Supports scheduled tasks, recurrent workflows that run automatically on your behalf - **Integrates with Anthropic's Skills framework:** structured, reusable task instructions that give Claude persistent operating context - Built with prompt injection detection and file deletion protection **Why it matters for NZ leaders** - Anthropic engineers now use Claude in 60% of their own work (up from 28% a year ago) and report 50% productivity gains. Claude Code's codebase is roughly 90% written by Claude itself. This is acceleration made visible. - 'Feeling less like back-and-forth communication and more like leaving messages for a colleague.' That framing is deliberate. This is not a tool you operate, it is a colleague you brief. - **The business implication:** one person producing the output of three to five people is no longer a thought experiment. It is Anthropic's daily reality. - **For NZ SMEs and professional services firms:** the cost-to-serve lever is being reset. Businesses that understand how to brief Cowork well will outcompete those that don't. **Attend the AI Toolkit Webinar** [#Let Claude Do the Work While You Think](https://academy.newzealand.ai/c/the-ai-toolkit-webinars/let-claude-do-the-work-while-you-think) **on Friday April 10th to see Claude Cowork in action.** ## **Microsoft Copilot Cowork** *- Enterprise-grade agentic execution in M365* **What it is** Copilot Cowork is Microsoft's enterprise answer to Claude Cowork, built in close collaboration with Anthropic and powered by the same Claude agentic framework. The key difference: where Claude Cowork runs locally on your device, Copilot Cowork runs in the cloud, inside your Microsoft 365 tenant, with full enterprise data protection. This is Wave 3 of Microsoft 365 Copilot, the move from 'chat that helps' to 'agent that acts.' It is powered by Work IQ: a contextual layer that grounds tasks in your emails, meetings, messages, files, and data across Outlook, Teams, Excel, and the full M365 suite. **What it does** - Delegates multi-step work across M365, you describe an outcome, Copilot creates a plan, executes it in the background, and checks in at clear approval points - **Calendar management:** reviews your schedule, flags conflicts, proposes and applies changes, including sending prep documents before meetings - **Meeting preparation:** pulls from emails, files, and past meetings to produce briefing documents, supporting analysis, and client-ready decks - Work IQ provides full organisational context, not just fragments of your data, but the whole picture - Runs within Microsoft's security, governance, identity, and compliance framework, enterprise-safe by default - Broad availability via the Frontier programme from late March 2026; general availability of Agent 365 from 1 May 2026 **Why it matters for NZ leaders** - If your organisation already runs Microsoft 365, this is the most immediately deployable agentic AI capability available today. There is no new infrastructure. No new vendor relationship. It runs inside what you already have. - **The collaboration with Anthropic is significant:** Microsoft is no longer betting exclusively on OpenAI. It is building a multi-model advantage. The best model for the job, regardless of who built it. - **For NZ enterprise and mid-market leaders:** your IT team and your operations managers need to be in a room this month discussing what Copilot Cowork means for your current M365 investment. - The operating question is not 'should we use this?' You are already paying for most of it. The question is 'how do we redesign our workflows to use it well?' ## **Perplexity Computer** *- Multi-model workflow orchestration* **What it is** Launched 25 February 2026, Perplexity Computer is the most ambitious leap in this briefing. It is not a single AI. It is an orchestration layer that coordinates 19 AI models simultaneously, including GPT, Claude, Gemini, and Grok, selecting the best model for each specific step in a workflow. It runs in an isolated compute environment with access to a real filesystem, a real browser, and real tool integrations. Tasks can run for hours. Or months. You can run dozens of Perplexity Computers in parallel, asynchronously, while you focus on other things. **What it does** - Automatically selects the right AI model for each sub-task (e.g. Gemini for deep research, Grok for speed, Claude for long-context recall) - Creates and coordinates sub-agents dynamically when it encounters a problem it cannot solve alone - Executes entire workflows, not just tasks, that can span hours or months without requiring ongoing input - Operates within an isolated compute environment with filesystem, browser, and tool access **Why it matters for NZ leaders** - This is what 'AI operating logic' looks like at its most advanced. You don't choose a model. You describe an outcome. The system routes intelligently. - **For NZ business leaders:** the relevant question is not 'can we afford this?' It is 'what workflows in our business currently take a week that this could execute in an afternoon?' - The Comet browser integration means Perplexity Computer can operate web-based workflows, filling forms, extracting data, navigating interfaces, that previously required a human at a keyboard. **Attend the AI Toolkit Webinar** [#Your AI Agent That Actually Does the Work](https://academy.newzealand.ai/c/the-ai-toolkit-webinars/the-ai-toolkit-your-ai-agent-that-actually-does-the-work) **on Perplexity Computer on March 20th.** ## **OpenClaw** *- Open-source personal AI agent* **What it is** OpenClaw is an open-source AI agent platform that runs on your own machine, laptop, server, or cloud instance. Built by Peter Steinberger and acquired by OpenAI in February 2026, it went from zero to 68,000 GitHub stars in 72 hours, making it one of the fastest-growing open-source projects in history. Think of it as a 24/7 AI worker that lives on your infrastructure. It connects to your files, your chat apps (WhatsApp, Slack, Teams, Discord), your browser, and over 50 third-party tools. It remembers everything across sessions. And it acts, autonomously, while you sleep. **What it does** - Runs persistently in the background, not just when you open a chat window - Maintains long-term memory of your preferences, projects, and past interactions via plain Markdown files (no black-box database) - **Executes multi-step tasks:** reading from one platform, processing, and acting across multiple systems in a single automated sequence - Supports 100+ community-built 'AgentSkills', modular capabilities it can autonomously extend - Isolates context per channel (WhatsApp conversations don't contaminate Slack context) - **Fully model-agnostic:** run on GPT, Claude, Gemini, or local models **Why it matters for NZ leaders** - OpenAI acquired it not for the AI capability, but for the workflow infrastructure. That is the signal: the race has moved from smarter models to smarter orchestration. - For NZ businesses, this is the DIY version of an enterprise AI platform, free, private, and powerful. The catch: it requires technical setup. - More importantly, the operating logic it represents, persistent, context-aware, multi-step AI execution, is the architecture every major platform is now racing to match. - Your people need to understand what context protocols are. OpenClaw is the most transparent demonstration of how they work. ## The Signal Behind All Four These four platforms are not four different bets. They are four different expressions of the same architectural shift. ***Prompting an AI and uploading your docs is now the manual way of using AI. The gap between those who know that and those who don't is widening every week.*** **The Real Risk** It is not that these tools will replace your people. It is that businesses whose people know how to use them will move at a pace yours cannot match. The constraint is not access. Every tool in this briefing is available to any New Zealand business today. The constraint is capability, specifically, whether your people understand the operating logic these platforms require. That means understanding: - What a context protocol is and why it matters - How to structure work so an agent can execute it without constant supervision - How to apply human judgement at the right checkpoints, not all checkpoints - How to measure whether AI execution is delivering against a business lever, not just producing output ## Three Decisions You Need to Make **01 - Which workflows will we redesign first?** Not which tools to buy. Which processes to rebuild around agentic execution. Name two. Assign an owner. Set a 90-day target metric. **02 - Who is accountable for AI capability uplift?** Learning to brief an agent is a skill. It is not the same as learning to prompt a chatbot. Your people need structured capability building. **03 - What are our guardrails?** Where does AI draft and a human approves? Where does AI execute and a human reviews? Where does a human decide, full stop? Write it down. Make it policy. The businesses getting this right are not doing more AI projects. They are redesigning how work happens, with AI in the operating logic from the start. That is a capability question, not a technology question. ## Power up your potential with practical AI skills. Contact us to discuss how Artificial Intelligence could boost your business. [Book an AI Chat](https://www.newzealand.ai/contact-us/?service=ai-chat#book) [Explore the AI Operating System](https://www.newzealand.ai/ai-os/) --- # Capability is cheap. Judgment is rare. Source: https://www.newzealand.ai/capability-is-cheap-judgment-is-rare/ 20 January 2026 · Best Practice # Curiosity is the antidote to obsolescence *These are the frameworks that will shape how we think, advise, and challenge New Zealand businesses this year.* **Justin Flitter** · Founder, AI New Zealand · 20 January 2026 · 4 min read Most businesses are treating AI like it's a shopping trip. They're comparing tools, collecting "use cases", and asking for demos like the next feature drop is going to save them. It won't. Not because AI isn't powerful, but because capability is becoming cheap. Everyone has access to the same models, the same interfaces, the same shiny "AI-powered" buttons. The differentiator is no longer access. It's thinking. This year, I'm anchoring everything I advise and teach around three big ideas. They're simple. They're uncomfortable. And if you get them right, they'll keep you relevant while the AI tide keeps rising. Most organisations don't have an AI problem. They have an incuriosity problem. They assume learning is what you do in a workshop once a year, then you get back to "real work". In an AI-driven world, that mindset doesn't just slow you down. It quietly makes you obsolete. Because the gap between thriving and struggling isn't determined by what you know today, it's determined by your hunger to understand what comes tomorrow. Curiosity isn't a personality trait. It's a survival mechanism. The businesses struggling right now aren't those without the right tools. They're the ones who've stopped asking good questions. They've trained their teams to be efficient, compliant, and busy, then act surprised when no one explores what's possible. Curiosity is cultivated. Deliberately. Systematically. Culturally. **What this looks like in practice:** - In meetings, replace "Any questions?" with "What are you curious about?" - Give teams permission to explore without needing a 30-page business case - Reward questions that reveal blind spots, not just answers that sound confident Because in 2026, the only truly obsolete skill is believing you already know enough. ## Critical thinking is your competitive moat AI commoditises answers. That's the part most leaders still haven't internalised. When anyone can generate a strategy doc, a proposal, a customer email sequence, and a board update in 90 seconds, "output" stops being impressive. The market gets flooded with plausible, polished, confidently-written mediocrity. And the risk isn't just sameness. The risk is confident wrongness at scale. Critical thinking is what stands between your organisation and AI theatre. It's the fusion of domain knowledge, lived experience, and intellectual courage - the willingness to challenge both AI outputs and conventional wisdom, then act decisively on what you learn. Human judgment becomes premium. And here's the uncomfortable truth: a lot of AI usage today is lazy. Copy-paste prompts. No context. No interrogation. No accountability. That's not innovation. That's outsourcing your thinking and calling it productivity. If you want a moat, build judgment. **What this looks like in practice:** - Make "Why?" a standard operating procedure: why this, why now, what assumptions? - Rotate a devil's advocate role so questioning becomes normal, not political - **Treat AI drafts as step one:** "AI draft human judgment discussion decision" AI can give you options. It can't give you conviction. That part is on you. ## Stop chasing use cases. Pull value levers. "Use cases" are where AI programs go to die. Not because they're wrong, but because they're usually backwards. Use cases describe what technology can do. Value levers define what the business must improve - revenue, cost-to-serve, cycle time, risk, customer experience. If you start with use cases, you end up with a long list of pilots and a short list of results. If you start with value levers, you don't "pilot AI". You move a number that matters. This is the shift I want more New Zealand leadership teams to make this year: stop asking "Where can we use AI?" and start asking "Which metric are we willing to be held accountable for, and what's the fastest AI-assisted path to move it?" Because the board doesn't fund curiosity forever. It funds outcomes. **What this looks like in practice:** - Pick 1–2 value levers only (not ten) and assign an owner - Set a baseline metric and a 90-day target - **Build an "AI value intake" gate:** every idea must name the lever, the metric, and the measurement plan You don't need more ideas. You need sharper priorities. ## How these three ideas connect Curiosity builds capability. Critical thinking deploys it with impact. Value levers make it real. That's the narrative for the year. Curiosity stops you becoming a dinosaur. Critical thinking stops you becoming a copycat. Value levers stop you becoming a science project. Together, they create the leadership advantage AI can't automate: judgment, context, and conviction. ## Clear next steps (do this in the next 30 days) If you want this to be more than a nice read and a motivating LinkedIn quote, here's the move. ## 1) Run a 60-minute "Value Lever Sprint" Get your leadership team in a room (or on a call). Decide: - The one lever you most need to move this quarter (cost-to-serve, sales cycle time, churn, risk, customer experience) - The baseline number - The 90-day target - The owner No tech talk until that's done. ## 2) Install a weekly curiosity habit Pick one: - A shared Curiosity Board where anyone can post a question worth exploring - 10 minutes in every team meeting for "What surprised us this week?" - **A standing rule:** "If AI is involved, someone must ask what we might be missing." Small rituals scale. Big speeches don't. ## 3) Create a "human judgment checkpoint" for AI work Make it policy that anything AI generates gets a quick human test: - What's the claim? - What evidence supports it? - What context is missing? - What could go wrong if we act on this? You're not slowing down. You're building a moat. This year, I'm not interested in AI hype, tool lists, or frantic experimentation with no scoreboard. I'm interested in helping New Zealand businesses think better, move faster, and stay relevant. If you want to build an organisation that doesn't just use AI, but wins with it, start here. - Curiosity is the antidote to obsolescence. - Critical thinking is your competitive moat. - Move a metric, not a use case. ## Power up your potential with practical AI skills. Contact us to discuss how Artificial Intelligence could boost your business. [Book an AI Chat](https://www.newzealand.ai/contact-us/?service=ai-chat#book) [Explore the AI Operating System](https://www.newzealand.ai/ai-os/) --- # CEO Briefing: AI Agents Are Doing Actual Work. Source: https://www.newzealand.ai/ceo-briefing-ai-agents-are-doing-actual-work/ 30 September 2025 · AI Agents # CEO Briefing: AI Agents Are Doing Actual Work. Something fundamental has shifted in the AI landscape. **Justin Flitter** · Founder, AI New Zealand · 30 September 2025 · 4 min read We're no longer talking about chatbots that answer questions or tools that help draft emails. We're talking about autonomous AI agents that can work independently for 30+ hours on complex projects, complete expert-level tasks that previously took human professionals 4-7 hours, and operate across multiple business functions without constant human oversight. This isn't incremental improvement. It's a threshold moment and New Zealand businesses are already experiencing the transformation firsthand. **The Evidence Is Compelling** OpenAI's latest testing reveals that advanced AI models now perform at near-human expert levels across finance, law, retail, and research. Anthropic's Claude Sonnet 4.5 has achieved 77.2% success rates on real-world software engineering benchmarks, tasks that require multi-step reasoning, tool usage, and self-correction. But here's what matters most: **82% of New Zealand businesses now use AI in some capacity, with 93% reporting improved worker productivity.** We're not just keeping pace with global AI adoption, we're leading it. The distinction is critical. These aren't better chatbots. They're autonomous agents capable of replicating complex academic research, generating comprehensive business documents, and managing entire workflows without human intervention at every step. **Why New Zealand Should Pay Attention** New Zealand is experiencing an AI productivity boom that's outpacing global averages. **70% of Kiwi CEOs report AI has made their workforce more efficient, compared to just 42% in Australia.** This isn't theoretical, it's happening now across sectors: - **Manufacturing:** Fisher & Paykel's Agentforce implementation is transforming production workflows - **Finance:** AI agents monitor transactions 24/7, delivering 74% reduction in manual audit efforts - **Compliance:** Companies report 63% cuts in compliance-related downtime The competitive implications are stark. Companies leveraging AI agents are becoming "Frontier Firms", organizations that outperform competitors and expand workforce capacity without proportional hiring. With New Zealand's tight labour market, AI agents offer a strategic solution to talent shortages while maintaining our productivity edge. We also have a regulatory advantage. New Zealand's "light touch, proportionate and risk-based" approach creates space for innovation while existing privacy and governance frameworks provide adequate oversight. This regulatory environment gives Kiwi businesses first-mover advantages over competitors in more restrictive jurisdictions. **The Opportunity Is Transformational** I see the opportunity across three dimensions: **Operational Excellence** AI agents can automate entire business processes, not just individual tasks. They're handling customer service inquiries, processing complex financial analysis, conducting research, and managing supply chain logistics autonomously. This represents a fundamental shift in how work gets done. **Strategic Advantage** Early adopters are creating competitive moats through AI-powered capabilities that competitors can't easily replicate. These aren't just efficiency gains, they're entirely new operating models that fundamentally change cost structures and service delivery. **Innovation Catalyst** AI agents excel at complex problem-solving and creative tasks. They can conduct comprehensive market research, generate strategic insights, and accelerate product development cycles. The technology is enabling companies to tackle challenges that were previously resource-prohibitive. **The Governance Challenge We Can't Ignore** The autonomous nature of AI agents introduces unprecedented governance challenges that require board-level attention. When systems can take thousands of actions daily without human review, traditional compliance approaches become insufficient. Operational risks multiply as agents initiate actions across multiple business functions simultaneously. Reputational risks escalate when AI interacts directly with customers and partners without human oversight. Financial risks compound when systems can commit organisational resources autonomously. New Zealand faces specific cybersecurity concerns, 28% of large businesses worry about AI-generated attacks, though only 6% of breaches have involved AI to date. Privacy risks are particularly acute given our robust Privacy Act 2020 requirements. Here's the reality: **69% of NZ executives see agentic AI as essential for coping with talent shortages and productivity pressures.** The question isn't whether to adopt AI agents, it's how to do so responsibly and strategically. **Questions Your Leadership Team Should Be Asking** **"What business processes could we completely reimagine if AI agents handled 60% of the workload?"** Move beyond task automation to process transformation. **"How do we establish governance frameworks for systems that operate autonomously across multiple functions?"** Traditional approval workflows are insufficient for real-time AI decisions. **"What new competitive advantages could we build through AI agent capabilities that competitors can't easily replicate?"** Focus on strategic differentiation, not just efficiency. **"How do we balance the productivity gains from AI agents with governance and risk management requirements?"** Develop embedded compliance that builds regulatory requirements directly into AI operations. **"What workforce implications arise when AI agents become permanent team members rather than tools?"** Plan for hybrid human-agent teams and new skill requirements. **"How do we prepare for the next wave of AI capabilities while maximizing current opportunities?"** Establish foundations for continuous AI evolution. **Your Path Forward** Immediate Actions (Next 30 Days) - **Conduct an AI audit** of current usage across your organization, many employees are already using AI tools informally - **Establish an AI governance council** with cross-functional leadership including legal, IT, HR, and key business units - **Assess your data infrastructure** and security posture for AI agent integration - **Begin pilot programs** with AI agents in low-risk, high-value processes **Medium-Term Priorities (Next 6 Months)** - **Develop comprehensive AI policies** aligned with New Zealand's Public Service AI Framework principles - **Invest in workforce training** for human-AI collaboration and establish new performance metrics for hybrid teams - **Partner with AI vendors** to implement enterprise-grade agent solutions with proper security and governance controls - **Create measurement frameworks** to track AI agent performance, risks, and business outcomes **The Window Is Now** The window for strategic AI agent adoption is now. Companies that move decisively will establish competitive advantages that become increasingly difficult for others to replicate. The technology is ready. The regulatory environment is supportive. And New Zealand businesses are already demonstrating global leadership in AI productivity gains. The question is: will you lead this transformation, or respond to it? *At AI New Zealand, we help business leaders navigate the strategic adoption of AI agents through unbiased advisory, practical skills-based training, and implementation support. If you're ready to explore what AI agents can do for your organization, let's talk.* ## Power up your potential with practical AI skills. Contact us to discuss how Artificial Intelligence could boost your business. [Book an AI Chat](https://www.newzealand.ai/contact-us/?service=ai-chat#book) [Explore the AI Operating System](https://www.newzealand.ai/ai-os/) --- # CEO Briefing: is Moltbot (formerly Clawdbot) safe at work? Source: https://www.newzealand.ai/ceo-briefing-what-is-clawdbot-now-moltbot-is-it-safe/ 2 February 2026 · Guidelines # CEO Briefing: What is Clawdbot (now Moltbot), is it safe to use at work? **OpenClaw / Moltbot / Clawdbot / Clawd Bot** **Justin Flitter** · Founder, AI New Zealand · 2 February 2026 · 4 min read A selfhosted personal AI "gateway" that you run on your own machine (often a Mac mini) and then talk to via WhatsApp, Telegram, Slack, etc. It routes your messages to various LLMs (Claude, ChatGPT, Gemini, local models), adds memory, notifications, and tool orchestration on top. OpenClaw is the latest proof that "AI that acts" is arriving faster than most organisations' governance can keep up. Treat them as signals, not solutions, and build your trust infrastructure before you build autonomy. This video is well worth a watch. OpenClaw is open-source software that turns AI from a reactive chatbot into an always-on agent that can take actions on your behalf, typically running on your own hardware and connecting through everyday messaging channels. Architecturally, it behaves like a local "gateway" between cloud models (for reasoning) and your private data (on the machine), often with persistent memory so it can carry context across sessions. The standout feature is "proactive" behaviour (a heartbeat) where the agent can wake itself up, monitor situations, and message you first, which meaningfully increases both usefulness and blast radius. **Why this matters to NZ CEOs** If an agent can read inboxes, touch calendars, access files, and send messages, it is no longer an IT experiment, it's operational delegation, which moves accountability straight to the CEO and board. From an NZ Privacy Act perspective, once personal information is involved you're expected to have safeguards that are "reasonable in the circumstances" to prevent loss, unauthorised access, misuse, or disclosure. NZ leaders should assume regulators and customers won't care that "the agent did it", they'll care whether your organisation put sensible controls around an autonomous system interacting with sensitive data. **The opportunity this presents** The upside is real: autonomous agents can compress coordination costs and cycle time by taking on repetitive digital work (triage, scheduling, follow-ups, summaries) across channels your teams already use. The strategic win isn't "cool automation", it's execution leverage: leaders can shift capacity from admin to customer value, while building repeatable workflows that scale across teams. Done well, this becomes a competitive advantage because AI commoditises answers, but organisations with strong judgement and guardrails can safely turn those answers into actions. **The impact on risk and governance** The core risk is prompt injection: untrusted content (emails, web pages, messages) can smuggle instructions that hijack the agent into leaking data or taking unintended actions. Security researchers increasingly frame "agent danger" as a design pattern problem: when an agent simultaneously, consumes untrusted inputs, has access to sensitive data/systems, and can take actions, the likelihood and impact of compromise spikes. Then there's commercial risk: autonomous loops can burn tokens and rack up unpredictable API costs, so CFO-grade spend controls need to be part of governance, not an afterthought. OpenClaw (AI-to-AI social posting) is more performance art than business capability, and the governance question is simple: are you paying to generate noise, or investing in systems that move a metric? OpenClaw previously called Clawdbot and Moltbot is a self-hosted AI assistant designed to run on your own machine (Mac/Windows/Linux) and operate through chat apps like WhatsApp, Telegram, Slack, Signal and iMessage. It's built for "agentic" work: reading and writing files, browser control (filling forms, extracting data), and even executing commands, with options to sandbox or restrict access. The official security guidance is blunt: there is no perfectly secure setup, so the goal is to deliberately limit who can trigger it, what it can touch, and where it can act. This is the "digital employee" moment: a tool that can take real actions inside your workflows, not just draft content, which means CEOs can finally link AI adoption directly to operational throughput and customer experience. The moment an AI agent can access customer emails, staff calendars, files, and internal systems, it becomes a Privacy Act issue as much as a productivity play, especially around safeguarding personal information. Because Moltbot can integrate across many services and store local transcripts, you need an NZ-appropriate governance stance on data handling, access control, and auditability, not just "let the IT team try it". **Questions CEOs should be asking** - Where would an autonomous agent *actually* move a value lever for us (revenue, cost-to-serve, cycle time, risk, customer experience), and what metric will we hold it to? - What are our "no-go zones" for agent access (customer PII, finance, payroll, production systems), and who signs off on exceptions? - How do we defend against indirect prompt injection when the agent reads untrusted content like emails and websites? - What actions require human approval (sending external messages, running code, purchasing, changing customer records), and how do we prevent approval fatigue? - What is our cost and usage control model (token budgets, spending caps, anomaly alerts) so an "agent loop" can't quietly turn into a finance incident? - If this agent causes a privacy breach, can we evidence "reasonable safeguards" and respond fast? **Steps CEOs should take next** Immediately (next 2–4 weeks): run a constrained pilot with read-only access wherever possible, strict scope, and clear human-in-the-loop gates for any irreversible action (external messaging, system changes, code execution). Immediately: implement "trust boundaries" as policy, treat all untrusted inputs as hostile, and assume prompt injection attempts will occur, because attackers can hide instructions in content the model reads. Medium term (30–90 days): create an agent governance standard (approved tools/connectors, access control model, logging and review, incident response playbook) aligned to NZ privacy expectations for reasonable safeguards. Medium term: only scale what moves a number, and kill anything that turns into "agent theatre" (including AI-to-AI content farms) unless it serves a defined customer, operational, or risk outcome. ## So is Moltbot safe to use at work? For most NZ organisations: These tools are **not safe for broad rollout**, and not safe for sensitive workflows on day one. For a controlled pilot with tight access controls, minimal permissions, and clear governance: **safe enough to trial**, especially for low-risk use cases like summarisation and internal drafting. Curiosity builds capability. Critical thinking deploys it with impact. In 2026, capability alone won't cut it, because everyone can spin up an agent. Judgement, governance, and disciplined value levers are the CEO edge AI can't automate. ## Power up your potential with practical AI skills. Contact us to discuss how Artificial Intelligence could boost your business. [Book an AI Chat](https://www.newzealand.ai/contact-us/?service=ai-chat#book) [Explore the AI Operating System](https://www.newzealand.ai/ai-os/) --- # ChatGPT vs Perplexity vs Claude vs Copilot: which wins? Source: https://www.newzealand.ai/chatgpt-perplexity-claude-m365-copilot-which-one/ 28 March 2025 · Insights # ChatGPT, Perplexity, Claude, M365 Copilot - Which one provided the best response to this advanced prompt? I've gave four AI assistants the same task, to prepare a detailed, board-level briefing for a New Zealand-based medical technology company preparing to launch and grow in North America. **Justin Flitter** · Founder, AI New Zealand · 28 March 2025 · 3 min read The specific models used: - ChatGPT 4.5 - Perplexity (Deep Research) - Claude 3.5 Haiku - Copilot for M365 I saved each response as a PDF, that I uploaded to Perplexity to evaluate and explain which gave the best response. **Here's the prompt that you can adjust to suit your situation.** You are a senior healthcare analyst with 20 years of experience advising global healthcare organisations on business process optimisation and technology adoption. You combine data-driven strategic thinking with deep empathy for patients and a clear understanding of frontline clinical realities. Prepare a detailed, board-level briefing for a New Zealand-based medical technology company preparing to launch and grow in North America. This company requires a robust understanding of the landscape to shape its go-to-market strategy. Your briefing must include: ## 1 Macro and Microeconomic Trends - Key healthcare spending trends in the US and Canada - Private vs public sector dynamics - Health insurance market shifts and payment model evolution (e.g., fee-for-service to value-based care) - Labour market pressures in healthcare (e.g., nursing shortages, burnout, unionisation) ## 2 Geo-Political Factors - Impact of upcoming elections and political sentiment on healthcare reform - Cross-border regulations and trade considerations for medical tech imports - Key legislation influencing procurement, privacy, and interoperability ## 3 Technology Landscape - Trends in AI, digital health, remote monitoring, EHRs, and data interoperability - Current adoption barriers and enablers in clinical and operational settings - Major players and potential partners or competitors ## 4 Strategic Recommendations - Market entry strategies (e.g., regional focus, partnership pathways, regulatory navigation) - Go-to-market positioning aligned with unmet needs of healthcare providers and patients - Operational readiness considerations for compliance, talent, and customer success - Suggestions for differentiating in a highly competitive innovation ecosystem Use data, case studies, and real-world examples where relevant. Maintain a tone that is strategic, empathetic, and commercially astute. **Here are the individual responses.** **Here's the evaluation from Perplexity (Reasoning with R1)** The three AI-generated briefings each offer distinct strengths in preparing a New Zealand medical technology company for North American market entry. Here's an evaluation of their approaches and effectiveness: ### 1. **Copilot Response (Search Result 1)** **Strengths**: - Provides **comprehensive regulatory detail**, including specific compliance requirements under the Drug Supply Chain Security Act and CMS Interoperability Rule[1] - Offers **clear operational guidance** with specific regional recommendations (California/Ontario) and partnership pathways[1] - Balances US/Canada comparisons effectively throughout all sections[1][3] **Weaknesses**: - Lacks forward-looking projections beyond 2024 - Less emphasis on workforce unionization trends emerging in North America - Limited discussion of political lobbying strategies --- ### 2. **ChatGPT Response (Search Result 2)** **Strengths**: - **Strong commercial focus** with specific partnership targets (Kaiser Permanente, UnitedHealthcare)[2] - Highlights **actionable market entry metrics** (35% value-based payments, 20% nurse vacancy rates)[2] - Provides **crisp competitive analysis** of major players and potential collaborators[2] **Weaknesses**: - Underdeveloped analysis of Canadian provincial differences - Minimal discussion of supply chain logistics - Limited treatment of intellectual property considerations --- ### 3. **Claude Response (Search Result 3)** **Strengths**: - Presents **most sophisticated economic modeling** ($6.2T US healthcare projection by 2028)[3] - Addresses **emerging labor dynamics** including unionization trends and workforce militancy[3] - Integrates **system-level efficiency analysis** with $200-500B savings potential[3] **Weaknesses**: - Less granular compliance detail than Copilot - No specific regional entry sequencing - Limited discussion of reimbursement mechanisms --- ### **Best Overall Response: Claude (Search Result 3)** **Rationale for Selection**: - 1. **Strategic Depth**: Projects market conditions through 2028 with detailed workforce and financing model analysis, critical for board-level planning[3] - 2. **Systemic Perspective**: Identifies $200-500B efficiency potential through care model innovation, aligning with MedTech value propositions[3] - 3. **Labor Insights**: Unique coverage of unionization trends and workforce militancy impacting implementation[3] - 4. **Financial Rigor**: Breaks down public/private financing ratios with hybrid model projections[3] While Copilot excels in regulatory detail and ChatGPT in commercial partnerships, Claude's combination of **long-term forecasting**, **labor trend analysis**, and **system-wide efficiency modeling** provides the most complete foundation for strategic decision-making. The briefing successfully bridges macroeconomic trends with operational realities, particularly valuable for a board assessing multi-year market entry commitments. **Key Improvement Opportunity**: All responses would benefit from deeper analysis of: - Intellectual property protection strategies - Indigenous health system considerations (particularly in Canada) - Post-market surveillance requirements - Comparative medical device approval timelines (FDA vs Health Canada) ## Power up your potential with practical AI skills. Contact us to discuss how Artificial Intelligence could boost your business. [Book an AI Chat](https://www.newzealand.ai/contact-us/?service=ai-chat#book) [Explore the AI Operating System](https://www.newzealand.ai/ai-os/) --- # Clarifying Trust Concerns With Popular AI Assistants Source: https://www.newzealand.ai/clarifying-trust-concerns-with-popular-ai-assistants/ 11 May 2025 · Guidelines # Clarifying Trust Concerns With Popular AI Assistants In this recent report, our collective concerns were verified in hard data. **Justin Flitter** · Founder, AI New Zealand · 11 May 2025 · 2 min read New Zealander's like many around the world don't trust AI tools. [@Peter Mangin](https://www.newzealand.ai/u/cbe9ba6a) - has shared some research I've been hearing a LOT of chatter about data security/privacy regarding LLMs recently. Generative AI has moved from the margins to the meeting room in record time. Yet one concern still tops every risk register: **"Will ChatGPT or Claude leak our sensitive data?"** Our latest New Zealandcentred study of enterprise editions says the fear is mostly misplaced, and the productivity upside is too big to ignore. As the majority seem to use ChatGPT and/or Claude, we pulled together this research: [https://aiinnovisory.com/llm-data-safety/](https://aiinnovisory.com/llm-data-safety/) [@Justin Flitter](https://www.newzealand.ai/u/d1450c78) Earlier this week we [published](https://academy.newzealand.ai/c/insights) the three misconceptions we get asked about at conferences - **Misconception 1:** "Everything shared with AI assistants becomes public" - **Misconception 2:** "All data uploaded to AI platforms is used to train models" - **Misconception 3:** "Private information becomes discoverable by others" [Read more](https://academy.newzealand.ai/c/insights) to learn How Major AI Platforms Actually Handle User Data [Frith Tweedie](https://www.linkedin.com/posts/frithtweedie_privacyweek-privacy-privacyonpurpose-activity-7326015856344268803-U0Ys?utm_source=share&utm_medium=member_desktop&rcm=ACoAAAIMIHgBtzkM-WFPP2ARmxuTYlHcaojRTEw) - posted this on LinkedIn A risk-based, right-sized approach to AI governance can give organisations the confidence to innovate without sleepwalking into legal or reputational trouble. Responsible AI isn't about saying no to innovation. It's about ensuring it's fair, lawful, and aligned with your organisation's values. In-house lawyers don't need to be AI experts, but they do play an important role in helping spot issues early and by asking smart questions. Want to recognise Privacy Week with your team but not had time to pull anything together? Simply Privacy is here to help with a FREE e-learning module. Privacy Week kicks off next week (12-16 May). We've created a free e-learning module on this year's theme of "Privacy on Purpose". Just five minutes long, it will encourage people to think about privacy whenever they're handling personal information. The module can be easily shared with your colleagues and teams to recognise the Privacy Week programme. Click here to access it here: [https://lnkd.in/g3w39tMs](https://lnkd.in/g3w39tMs) You can also access last year's version on "Busting Privacy Myths. We hope you find it useful The full details Privacy Week 2025 programme from the [Office of the Privacy Commissioner, New Zealand](https://www.linkedin.com/company/office-of-the-privacy-commissioner/) is available here: [https://lnkd.in/gwbPXBC5](https://lnkd.in/gwbPXBC5) [@Anchali Anandanayagam](https://www.newzealand.ai/u/60590653) from Hudson Gavin Martin Has published an [AI Policy Sheet-Sheet](https://www.linkedin.com/pulse/tips-from-trenches-your-ai-policy-cheat-sheet-hudson-gavin-martin-pupgc/?trackingId=0NrMIhsSRemIveKhoFHqEg%3D%3D) And have open-sourced their Principals Based Policy framework. Got Questions? Ask them here and we'll get an expert to answer them. ## Power up your potential with practical AI skills. Contact us to discuss how Artificial Intelligence could boost your business. [Book an AI Chat](https://www.newzealand.ai/contact-us/?service=ai-chat#book) [Explore the AI Operating System](https://www.newzealand.ai/ai-os/) --- # Dropping this AI Prompt Guide, for us to refer back to Source: https://www.newzealand.ai/dropping-this-ai-prompt-guide-for-us-to-refer-back-to/ 23 April 2025 · Prompts # Core Principles Anthropic's released an [hour long deep dive](https://www.youtube.com/watch?v=T9aRN5JkmL8) on prompt engineering. **Justin Flitter** · Founder, AI New Zealand · 23 April 2025 · 2 min read It covers: - What makes a good prompt - Handling ambiguity, reasoning paths, and edge cases - Enterprise vs research prompting strategies - Using personas, metaphors, and structured logic - Jailbreaking, trust, and testing prompts at scale OpenAI published also recently published a comprehensive [Prompting Guide](https://cookbook.openai.com/examples/gpt4-1_prompting_guide). They are both rather technical. So I asked Perplexity to watch the video and read the guidelines and create a AI Prompting Playbook anyone can use. Here's a non-technical guidebook synthesizing insights from OpenAI's GPT-4.1 prompting guide and Anthropic's expert discussions: - --- - **AI Prompting Mastery: A Practical Guide** Modern AI systems like ChatGPT and Claude require thoughtful communication. This guide combines OpenAI's latest recommendations with Anthropic's field-tested strategies. ## 1 Precision Beats Poetry ChatGPT responds best to literal, unambiguous instructions. Instead of "Make this sound fancy," try "Rewrite this paragraph at 12th-grade reading level using business formal tone." ## 2 Steerability These models course-correct instantly with clear feedback. If responses miss the mark, add one-sentence guidance like: *"Focus only on the environmental impacts listed in paragraph three"* ## 3 Structured Thinking Anthropic's experts recommend "pre-building" the model's reasoning path: *"First analyze the customer's main complaint, then suggest three solutions from our policy handbook, finally format as bullet points."* ## Practical Techniques **The 4-Part Prompt Formula** ## 1 Role Assignment: "You're a veteran science teacher explaining to 13-year-olds" ## 2 Task Context: "We're studying planetary orbits in our solar system" ## 3 Specific Request: "Create a 300-word story about Mercury's day-night cycle" ## 4 Constraints: "Use 2 analogies, avoid technical terms >6 letters" **Comparison: Old vs New Approaches** **Common Pitfalls & Fixes** - **Hallucinations**: Add "Base responses strictly on the attached document" - **Overly Creative**: Use "Provide only verified facts from credible sources" - **Format Errors**: Specify "Use Markdown headers and bullet points" ## Advanced Applications **Enterprise Use Cases** - **Legal teams:** "Compare sections 4.2-4.5 of both contracts. Highlight discrepancies in liability clauses using red text." - **Customer service:** "Identify frustrated customers by analyzing email sentiment. Suggest three de-escalation phrases from our playbook. **Creative Brainstorming** *"Generate 10 marketing taglines for eco-friendly shoes. Then select the top 3 based on: memorability, emotional appeal, brand alignment."* ## Future-Proof Strategies ## 1 Iterative Testing: Start simple, then layer constraints ## 2 Persona Crafting: "You're a skeptical editor fact-checking this article" 3. **Feedback Loops**: After initial output, add "Now remove industry jargon and add two case examples" --- **Key Takeaway**: Modern AI responds best to explicit, structured guidance that pre-answers the "why, who, and how" of your request. Treat prompting like coaching an exceptionally capable intern - clear direction yields extraordinary results. ## Power up your potential with practical AI skills. Contact us to discuss how Artificial Intelligence could boost your business. [Book an AI Chat](https://www.newzealand.ai/contact-us/?service=ai-chat#book) [Explore the AI Operating System](https://www.newzealand.ai/ai-os/) --- # Essential playbook of 60 AI task-boosters Source: https://www.newzealand.ai/essential-playbook-of-60-ai-task-boosters/ 15 December 2025 · Guidelines # Email & Communication **Justin Flitter** · Founder, AI New Zealand · 15 December 2025 · 14 min read Here's our comprehensive collection of **60 practical how-to guides** for using AI tools like ChatGPT, Gemini, Claude, and Perplexity to complete common daily work tasks. These are designed to be actionable guides that you can implement immediately. ## 1. **Write Professional Emails** Use ChatGPT, Claude, or Gemini to draft, rewrite, and polish business emails quickly. - **How to do it:** Provide context (recipient, purpose, tone), paste key points or existing draft, request specific style (formal, friendly, concise), then review and personalise before sending. - **Pro tip:** Use the C.R.E.A.T.E. Framework: Context, Role, Examples, Action, Tone, and Emphasis for better results ## 2. **Summarise Email Threads** Quickly understand long email conversations and extract action items. - **How to do it:** Copy the entire email thread, paste into your AI tool, and ask: "Summarise this email thread and highlight action items". - **Time saved:** Reduces 30-minute email reviews to under 2 minutes. ## 3. **Generate Email Subject Lines** Create compelling subject lines that increase open rates. - **How to do it:** Describe your email content and goal, ask for 5-10 subject line variations, request A/B testing options with different tones. - **Bonus:** Request versions with emojis for casual communications. ## 4. **Automate Email Responses** Set up AI-powered email drafts via Gmail/Outlook integration. - **How to do it:** Connect ChatGPT to your email via connectors to Gmail or Outlook, create templates for common inquiries, set up automated draft creation, then review and send approved drafts. - **Result:** Handle routine emails 4x faster while maintaining personalisation. ## Meetings & Collaboration ## 5. **Transcribe and Summarise Meetings** Use [Otter.ai](http://otter.ai/) to automatically capture meeting notes and action items. - **How to do it:** Connect Otter to Zoom, Teams, or Google Meet, let AI transcribe in real-time (up to 95% accuracy), review the automated summary and action items, then share the transcript with your team. - **Benefit:** Never miss important details and get meeting summaries in 30 seconds instead of manual note-taking. ## 6. **Schedule Meetings Automatically** Use AI scheduling assistants to eliminate back-and-forth emails. - **How to do it:** Set up CalendarBridge, Clockwise, or Motion, connect all your calendars (Google, Outlook, Apple), CC the AI assistant on scheduling emails, and let AI propose and book optimal times. - **Impact:** Reduce scheduling time by 75% and eliminate double-bookings. ## 7. **Prepare Meeting Agendas** Create structured meeting agendas quickly with AI assistance. - **How to do it:** Tell AI the meeting purpose and attendees, request an agenda with time allocations, ask for discussion questions, then customise and share with your team. - **Enhancement:** Include pre-reading materials and expected outcomes for each agenda item. ## 8. **Generate Meeting Follow-ups** Create action-focused meeting summaries automatically. - **How to do it:** Upload meeting transcript or notes to ChatGPT or Claude, ask AI to extract action items with owners, request timeline for deliverables, then send as follow-up email. - **Outcome:** Ensure accountability and clarity on next steps. ## Writing & Content Creation ## 9. **Write Reports and Documentation** Create professional business documents faster without starting from scratch. - **How to do it:** Outline key sections and objectives, provide data, findings, or research, ask AI to draft each section, then edit for accuracy and company voice. - **Speed:** Complete first drafts in 1/3 of the usual time. ## 10. **Proofread and Edit Documents** Use AI to catch errors and improve clarity instantly. - **How to do it:** Paste text into ChatGPT, Claude, or Wordvice AI, request grammar, spelling, and style checks, ask for clarity and flow improvements, then review suggestions and apply changes. - **Options:** Choose intensity levels from light proofreading to intensive editing with restructuring. ## 11. **Create Social Media Content** Generate posts for multiple platforms with platform-specific formatting. - **How to do it:** Describe your message and audience, specify platform (LinkedIn, Twitter, Instagram), request 3-5 variations, then schedule using Buffer, SocialBee, or Plann. - **Efficiency gain:** Create a week's content in 30 minutes ## 12. **Write Blog Posts and Articles** Draft long-form content with AI assistance while maintaining your voice. - **How to do it:** Provide topic and target audience, request outline with key sections, have AI expand each section, then add personal insights and examples. - **Best practice:** Always fact-check AI-generated content and add your unique perspective. ## 13. **Rewrite Content for Different Audiences** Adapt existing content for executives, technical teams, or general audiences. - **How to do it:** Paste original content, specify new audience, request appropriate tone and complexity level, then compare versions and select the best. - **Use case:** Transform technical reports into executive summaries or vice versa. ## Data & Analysis ## 14. **Analyse Spreadsheet Data** Use Copilot in Excel or ChatGPT to extract insights from data. - **How to do it:** Upload spreadsheet or describe data, ask specific questions about trends or patterns, request visualisations or pivot tables, then use AI-generated formulas for calculations. - **Capability:** Analyse complex datasets without advanced Excel skills ## 15. **Create Excel Formulas** Generate complex formulas from plain English descriptions - **How to do it:** Describe what you need to calculate, use ChatGPT, Copilot, or Excel Formula Bot, ask for formula explanation, then test and adjust as needed. - **Examples:** Create VLOOKUP, XLOOKUP, SUMIFS, or nested IF statements instantly ## 16. **Clean and Format Data** Automate tedious data cleaning tasks - **How to do it:** Identify data quality issues (duplicates, formatting inconsistencies), ask AI for cleaning steps or formulas, request standardisation rules, then apply and verify results. - **Time savings:** Reduce hours of manual data cleaning to minutes. ## 17. **Generate Data Visualisations** Create charts and graphs that tell compelling stories. - **How to do it:** Upload or describe your dataset, specify visualisation type needed, ask AI to recommend best chart type for your data, then export and customise in your tool. - **Smart feature:** AI can identify which chart type best represents your data patterns. ## Presentations & Design ## 18. **Create Presentation Slides** Use Gamma or Canva AI to generate full presentation decks - **How to do it:** Input your topic and key points, let AI generate a complete slide deck, review and customise design, then add branding and images. - **Speed:** Create professional presentations in minutes instead of hours. ## 19. **Design Visuals and Graphics** Create professional graphics without design skills. - **How to do it:** Describe visual concept or purpose, use Canva's Magic Design or AI image generators, request specific style or brand colours, then download and use in materials - **Accessibility:** Democratises design for non-designers. ## 20. **Generate Presentation Talking Points** Create speaker notes and presentation scripts - **How to do it:** Paste slide content into ChatGPT or Claude, ask for elaboration on each point, request examples or analogies, then practice with the AI-generated script. - **Enhancement:** Ask AI to anticipate audience questions and prepare responses. ## Research & Learning ## 21. **Conduct Literature Reviews** Use Perplexity, Elicit, or ResearchRabbit to accelerate research. - **How to do it:** Input research question or topic, let AI find relevant papers and sources across 125+ million papers, review AI-generated summaries, then export citations in required format. - **Impact:** Complete literature reviews in hours instead of weeks. ## 22. **Summarise Long Documents** Extract key points from PDFs, reports, and lengthy documents. - **How to do it:** Upload document to ChatGPT, Claude, Adobe Acrobat AI, or Sharly, ask for executive summary or specific sections, request key insights and recommendations - **Feature:** Get numbered citations linking back to source material for verification. ## 23. **Learn New Skills Quickly** Use AI as a personal tutor that never gets tired. - **How to do it:** Enable Study Mode in ChatGPT or Claude, ask AI to explain concepts step-by-step, request practice problems or exercises, then get feedback on your understanding. - **Advantage:** AI guides you through problems rather than just giving answers. ## 24. **Translate Documents** Convert content to multiple languages professionally. - **How to do it:** Paste text or upload document, use DeepL, Google Translate, or ChatGPT, specify target language and formality level, then review for context and accuracy. - **Best for:** DeepL excels at European languages with nuanced translations. ## Project Management ## 25. **Break Down Complex Projects** Create detailed project plans automatically. - **How to do it:** Describe project goals and constraints, ask AI to create task breakdown with dependencies, request timeline and milestones, then export to Asana, Notion, or Excel. - **Benefit:** Identify potential risks and bottlenecks before they occur. ## 26. **Automate Task Scheduling** Use Motion or Trevor AI to optimally time-block your day - **How to do it:** List all tasks with time estimates, connect calendar accounts, let AI optimise daily schedule based on your working patterns, then review and approve the plan. - **Result:** Increase task completion rates from 40% to 85%. ## 27. **Track Project Progress** Generate status reports automatically for stakeholders. - **How to do it:** Input project milestones and completion status, ask AI to create progress summary with insights, identify risks and blockers, then share with stakeholders. - **Automation:** Set up automated weekly reports in project management tools. ## 28. **Manage Team Workload** Balance tasks across team members effectively. - **How to do it:** List team capacity and current assignments, ask AI to recommend task distribution, identify overallocation risks, then adjust based on AI suggestions. - **Insight:** AI can predict burnout risks based on workload patterns. ## Customer Service ## 29. **Set Up AI Chatbots** Automate customer support responses 24/7. - **How to do it:** Define common customer questions, train chatbot with FAQs and approved responses, set escalation rules for complex issues, then monitor and refine responses. - **ROI:** Handle 60% of inquiries instantly while reducing support costs. ## 30. **Draft Customer Responses** Reply to inquiries professionally and empathetically. - **How to do it:** Paste customer message, provide relevant context or company policy, request empathetic, solution-focused response, then personalise before sending. - **Speed:** Respond to customer queries 4x faster. ## 31. **Handle Customer Complaints** Craft diplomatic responses to negative feedback. - **How to do it:** Share complaint details with AI, ask for empathetic acknowledgment, request solution-oriented language, then add a personal touch before sending. - **Outcome:** Improve customer satisfaction scores and reduce escalations. ## Marketing & Sales ## 32. **Generate Marketing Copy** Create ads, taglines, and promotional content at scale. - **How to do it:** Describe product and target audience, request multiple copy variations, specify channel and character limits, then test and refine top options. - **Volume:** Generate 50+ variations in minutes for A/B testing. ## 33. **Write Cold Outreach Emails** Create personalised sales emails that convert. - **How to do it:** Research prospect background, provide context to AI about their pain points, request warm, value-focused message, then customise with specific details. - **Best practice:** Use the "hook-benefit-CTA" structure for higher response rates. ## 34. **Brainstorm Campaign Ideas** Generate creative marketing concepts quickly - **How to do it:** Share campaign goals and target audience, ask for 10-15 creative concepts, request different approaches (emotional, logical, humorous), then develop promising ideas further. - **Innovation:** AI can suggest unexpected connections and fresh perspectives. ## 35. **Analyse Competitor Content** Extract insights from competitor materials systematically - **How to do it:** Gather competitor content (websites, ads, posts), ask AI to identify patterns and themes, request gap analysis, then develop differentiation strategy. - **Strategic value:** Identify market opportunities others have missed. ## HR & Recruitment ## 36. **Write Job Descriptions** Create compelling, inclusive job posts that attract talent. - **How to do it:** Provide role title and key requirements, request structured job description, ask for inclusive language review, then add company-specific details. - **Inclusion:** AI can identify biased language and suggest neutral alternatives. ## 37. **Screen Resumes** Quickly evaluate candidate fit against requirements. - **How to do it:** Upload resume and job description to AI, ask to assess qualification match percentage, request skills gap analysis, then create shortlist with rationale. - **Efficiency:** Review 100+ resumes in the time it takes to manually review 10. ## 38. **Prepare Interview Questions** Generate role-specific, behavioural interview questions. - **How to do it:** Describe role and key competencies needed, request both behavioural and technical questions, ask for evaluation criteria, then customise for your company culture. - **Structure:** Get STAR method questions automatically formatted. ## 39. **Practice Job Interviews** Prepare for interviews with AI role-play. - **How to do it:** Provide job description to AI, ask AI to conduct mock interview as the hiring manager, answer questions in real-time, then request detailed feedback and improvement tips - **Benefit:** Build confidence and refine answers before the real interview. ## Personal Productivity ## 40. **Create Daily Plans** Use AI to optimise your schedule and priorities. - **How to do it:** List all tasks and appointments, use Trevor AI, Morgen, or Motion for AI planning, let AI suggest optimal time blocks based on your energy levels, then review and adjust as needed. - **Smart feature:** AI learns your working patterns and adapts over time. ## 41. **Prioritise Tasks** Identify what truly matters using AI-powered analysis. - **How to do it:** List all pending tasks with context, ask AI to rank by urgency and impact using frameworks like Eisenhower Matrix, request reasoning for priorities, then focus on top-ranked items. - **Clarity:** Reduce decision fatigue and focus on high-impact work. ## 42. **Set Goals and Track Progress** Create SMART goals with AI guidance and accountability. - **How to do it:** Share broad objectives with AI, ask to create specific, measurable, achievable, relevant, and time-bound goals, request milestones and metrics, then schedule regular check-ins for progress. - **Motivation:** AI provides coaching and celebrates wins ## 43. **Overcome Decision Paralysis** Use AI for structured decision analysis. - **How to do it:** Describe your decision and available options, ask for comprehensive pros/cons analysis, request risk assessment for each option, then use insights to inform your choice. - **Framework:** Apply decision-making frameworks like weighted scoring. ## Finance & Administration ## 44. **Create Invoices and Quotes** Generate professional financial documents quickly. - **How to do it:** Provide service/product details and client information, request invoice or quote format with payment terms, add company-specific conditions, then save as template for future reuse. - **Automation:** Connect to accounting software for automatic record-keeping. ## 45. **Analyse Financial Data** Extract insights from financial reports and statements. - **How to do it:** Upload financial statements or data, ask specific questions about performance trends, request variance analysis and forecasting, then identify areas for cost reduction or growth. - **Insight:** Spot financial patterns humans might miss. ## 46. **Automate Expense Reporting** Simplify expense tracking and reporting processes. - **How to do it:** Take photos of receipts (AI extracts data), upload to accounting software like Xero or QuickBooks, let AI categorise expenses by type, then review and submit reports. - **Time saved:** Reduce expense reporting from hours to minutes monthly. ## Legal & Compliance ## 47. **Review Contracts** Use AI to identify risks in legal agreements. - **How to do it:** Upload contract to goHeather, LegalFly, or similar AI contract review tool, request comprehensive risk assessment, review flagged clauses and suggested redlines, then consult lawyer for critical issues. - **Coverage:** AI can spot risky clauses in thousands of contract types. ## 48. **Draft Standard Agreements** Create NDAs and basic contracts from templates. - **How to do it:** Specify agreement type and parties involved, provide key terms and conditions, let AI generate draft based on standard templates, then have legal professional review before execution. - **Cost savings:** Reduce legal fees for routine agreements. ## 49. **Ensure Regulatory Compliance** Check documents against industry regulations automatically. - **How to do it:** Describe your industry and applicable regulations, upload documents for compliance review, ask AI to flag potential compliance issues, then implement suggested changes and document. - **Protection:** Proactive compliance monitoring reduces legal risks. ## Creative & Video ## 50. **Create and Edit Videos** Use AI for efficient video production. - **How to do it:** Upload raw footage to Descript, Gling, or OpusClip, let AI automatically remove silences, filler words, and bad takes, add AI-generated captions and chapters, then export polished video. - **Time savings:** Cut editing time by 70%. ## 51. **Generate Video Scripts** Write engaging video content with strong hooks. - **How to do it:** Describe video topic, target audience, and desired length, request script with attention-grabbing hook and clear CTA, ask for timing and scene suggestions, then practice and refine delivery. - **Optimisation:** AI can suggest titles and thumbnails for YouTube success. ## 52. **Create Social Media Clips** Turn long videos into short, viral-worthy clips. - **How to do it:** Upload full video to OpusClip or similar tool, let AI identify the most engaging moments using virality algorithms, review and select AI-generated clips, then add branding and publish - **Versatility:** Works for podcasts, interviews, tutorials, gaming, and vlogs ## Coding & Development ## 53. **Generate Code Snippets** Write code from natural language descriptions. - **How to do it:** Describe what the code should accomplish, use ChatGPT, Claude, or GitHub Copilot, request specific programming language or framework, then test and integrate into your project. - **Productivity:** Developers code 126% more projects weekly with AI assistance. ## 54. **Debug Code Errors** Find and fix bugs faster with AI assistance. - **How to do it:** Paste error message and relevant code context, ask AI to identify root cause and explain the issue, request fix suggestions with explanations, then test and validate the solution. - **Accuracy:** AI can catch edge cases and potential regressions. ## 55. **Write Unit Tests** Generate comprehensive test cases automatically. - **How to do it:** Provide function or code module to test, ask for comprehensive test coverage including edge cases, request specific testing framework format, then integrate into your test suite. - **Coverage:** Ensure robust testing without tedious manual test writing. ## 56. **Explain Complex Code** Understand unfamiliar codebases quickly. - **How to do it:** Paste code snippet or function, ask for line-by-line explanation, request clarification on purpose and logic flow, then document insights for future reference. - **Learning:** Accelerate onboarding and knowledge transfer. ## Brainstorming & Innovation ## 57. **Generate Business Ideas** Explore new opportunities with AI creativity - **How to do it:** Describe your market, skills, interests, or constraints, ask for 20-50 diverse business concepts, request evaluation criteria (feasibility, profitability, scalability), then develop the most promising ideas. - **Innovation:** AI can suggest combinations you wouldn't have considered ## 58. **Solve Problems Creatively** Find innovative solutions to business challenges. - **How to do it:** Clearly describe the problem and constraints, ask for diverse solution approaches from different angles, request unconventional or lateral thinking options, then evaluate feasibility with AI's help. - **Method:** Use techniques like "reverse brainstorming" or "SCAMPER" with AI. ## 59. **Conduct SWOT Analysis** Analyse strengths, weaknesses, opportunities, and threats systematically. - **How to do it:** Describe your business, project, or initiative, request comprehensive SWOT analysis with data-backed insights, ask for specific examples and evidence, then use for strategic planning. - **Depth:** AI can provide industry context and competitive intelligence. ## 60. **Role-Play Scenarios** Practice difficult conversations safely. - **How to do it:** Describe scenario (negotiation, conflict, pitch) and assign role to AI, conduct the conversation with AI playing the other party, try different approaches and strategies, then get feedback on communication effectiveness. - **Confidence:** Build skills without real-world consequences. ## Getting Started Tips - **Start small:** Pick 2-3 guides most relevant to your daily work and master them first. - **Experiment with models:** Different AI tools excel at different tasks - ChatGPT for versatility, Claude for nuanced writing, Gemini for Google Workspace integration, Perplexity for research. - **Refine your prompts:** Be specific about context, desired format, tone, and length. The clearer your instructions, the better the output. - **Always verify:** AI can make mistakes or "hallucinate" information. Fact-check important content and add your expertise. - **Maintain your voice:** Use AI as a starting point, then personalise and add your unique insights. - **Protect sensitive data:** Be cautious about uploading confidential information to AI tools. Use enterprise versions with proper security when handling sensitive business data. ## Key Productivity Gains Research shows that workers using AI tools effectively can experience: - **59% increase in output** for business tasks - **126% more projects completed** for programmers - **30% improvement** in customer satisfaction when using chatbots - **20-30% efficiency boost** in project work - **58% reduction** in repetitive manual tasks The key is consistent use and continuous learning. Start implementing these guides today, and you'll quickly discover which tools deliver the most value for your specific work. The future of work isn't about AI replacing humans - it's about humans augmented by AI working smarter, faster, and more creatively than ever before. ## Power up your potential with practical AI skills. Contact us to discuss how Artificial Intelligence could boost your business. [Book an AI Chat](https://www.newzealand.ai/contact-us/?service=ai-chat#book) [Explore the AI Operating System](https://www.newzealand.ai/ai-os/) --- # Five lessons from the Great New Zealand AI Roadshow Source: https://www.newzealand.ai/five-things-we-have-learnt-so-far-on-the-great-new/ 7 June 2026 · Guidelines # 1. The gap is not adoption. It is the operating model. We are partway through the [Great New Zealand AI Roadshow](https://www.newzealand.ai/roadshow/), and the rooms have been full, honest, and a little uncomfortable in the best way. **Justin Flitter** · Founder, AI New Zealand · 7 June 2026 · 4 min read Nine sessions a day. Founders, lawyers, security leaders, mentors and marketers, all on one stage, all pointed at the same question: where are Kiwi businesses actually at with AI, not where the vendor decks say we should be? Here is what the rooms have taught us so far. New Zealand punches above its weight on AI. Claude usage here in 2025 ran roughly four times the global per-capita average. Seventy-eight percent of businesses have touched the tools. But only twelve percent have moved from experimentation to genuine strategic integration. That gap has a name. AI theatre. Better screenshots, same bottlenecks. The promise of transformation without the operating change underneath it. Asa Cox from Pneuma put hard numbers behind the same idea. After running adoption work with more than 100 NZ businesses, his read is that the technology can already do about 95 percent of what most businesses want. The remaining gap is human: culture, change, communication, training. Globally, around 70 percent of AI projects still fail, and almost none of them fail on the technology. The honest takeaway: you do not need to wait for better models. You need to build the capability that lets your team absorb what already exists. ## 2. Your data is not leaking. It is being handed over. Raymond Schippers from Check Point flipped the security story on its head. The old threat model assumed attackers were trying to take your data. The new reality is that your best people are quietly giving it away, in exchange for being a bit faster. Seventy-eight percent of organisations are using AI. Forty-seven percent of them have no AI-specific security control in place at all. Anchali Anandanayagam from Hudson Gavin Martin made the legal version of the point. When Air Canada's chatbot invented a refund policy, the airline argued the bot was a separate legal entity. The court said no. You own what your AI says. Raymond's sharpest line is the one for every executive: prompts are not controls. Telling an agent not to delete the database is like asking a teenager to clean their room. It might work. It might not. Real controls are technical: least privilege, human review on high-risk actions, rollback. Not instructions. ## 3. Train your AI to fight you. AI is trained to be a lap dog. You give it a half-baked idea, it hands back a five-star rating. Steve Ballantyne nailed it on the Auckland stage: a junior analyst with a lot of confidence and not enough context, telling you exactly what you want to hear. That is why strategy decks are starting to sound the same. His fix is the one we keep coming back to. Build a custom GPT and instruct it to be contrarian on purpose. Challenge every premise. Steelman the opposite position. Surface what you are avoiding. Run your next board paper through it before you ship, and keep only what survives the fight. AI commoditises answers. Critical thinking is your competitive moat. Human judgement becomes premium. ## 4. Do not automate a broken workflow. Richard Kenyon from Datacom drew the line that most businesses miss. Adoption is using AI to do today's work faster. Adaptation is redesigning the work itself. Most of the value lives in adaptation. Most of the effort goes into adoption, bolting AI onto workflows that were designed around legacy tech and legacy assumptions. AI does not fix a broken process. It accelerates it. So the strategy question is not which tool. It is how you want the business to sense, think, decide and act when AI is in the loop. ## 5. The price of creativity has collapsed. After a day of governance and risk, Steve Ballantyne reminded the room that AI is also one of the most powerful creative tools ever built, and most organisations only use it for efficiency. Brand mascots that used to need a Disney budget. A piping company shipping a Springsteen-style brand anthem from a folder of old photos. A full concept ad built end to end in 90 minutes, the kind of work that used to be a $20k to $50k cheque before production even started. The barrier to creative expression has basically disappeared. Most businesses do not have a shortage of ideas. They have a shortage of bandwidth to test them. AI fixes that. ## The thread that runs through all of it One person being great at prompting is not a strategy. The businesses pulling ahead are not using better AI. They are feeding it better context, and they are encoding that context once: methodology, brand voice and workflows, connected to real systems, used by everyone, improving every month. That is not an adoption problem. It is an operating model problem. And the rooms on this roadshow are full of people who have just worked that out. The tools are ready. The question is whether you are. If you want help figuring out where AI will actually move the needle in your business, come find us at [newzealand.ai](https://www.newzealand.ai/). ## Power up your potential with practical AI skills. Contact us to discuss how Artificial Intelligence could boost your business. [Book an AI Chat](https://www.newzealand.ai/contact-us/?service=ai-chat#book) [Explore the AI Operating System](https://www.newzealand.ai/ai-os/) --- # Stop Dabbling: The AI Muscle Gap holding back New Zealand. Source: https://www.newzealand.ai/stop-dabbling-the-ai-muscle-gap-holding-back-new/ 28 January 2026 · Guidelines # Stop Dabbling: The AI Muscle Gap holding back New Zealand. New Zealand's AI journey is not primarily a technology story. It is a maturity story, and an increasingly urgent one. **Justin Flitter** · Founder, AI New Zealand · 28 January 2026 · 9 min read The tools are racing ahead. Most organisations are not. Across hundreds of conversations with individuals and businesses throughout the country, and informed by recent research into how AI supports complex problem-solving and organisational knowing, a clear pattern has emerged: we already have access to world class AI models, but only a small minority are building the organisational muscles to use them wisely, safely, and distinctively. What follows is a concise look at the most important dimensions of AI maturity in the New Zealand context, and the divide that is opening up between those who dabble and those who operationalise. ## 1. Most Kiwi Organisations Are Still AI-Immature Beneath the hype, AI maturity in New Zealand clusters at the bottom of the curve. Among larger organisations, around two thirds now report using some form of AI, up sharply over the last few years. For small and medium enterprises, the picture is very different. Around 68 percent report they have no plans to evaluate or invest in AI at all. From the ground, this looks like: - **Level 0–1: Ad hoc and fragile** - No AI strategy, minimal or no governance, and experiments that live in the heads of a few enthusiasts. When those people move on, the capability disappears with them. - **Level 2: Emerging structure** - Tools such as Copilot and ChatGPT are available, there are pockets of usage, but AI is not embedded into core workflows. Training is thin, and there is no shared view of what "good" looks like. Very few organisations have moved into the stages where AI is supported by shared infrastructure, clear governance and leadership sponsorship, and treated as an ongoing organisational capability rather than a one off initiative. More organisations are now starting. Very few are yet reshaping their operating model or identity through AI. ## 2. The Hard Problems Are Human and Structural, Not Technical The main adoption barriers are not about models or compute. They are about people, process, and the particular shape of New Zealand business. **Micro businesses and family ownership** New Zealand is a nation of very small firms. Approximately 97 percent of businesses have fewer than 20 employees, around 70 percent are sole traders, and roughly 90 percent have five or fewer staff. A significant share are family owned, with most equity held by family shareholders. That structure tends to create: - Strong relationships, community roots, and long term thinking - Reasonable basic technology - Weakly documented processes and informal governance Work often lives in people's heads and in personal networks, not in systems or process maps. Without a clear view of how work actually gets done today, it is extremely hard to safely redesign that work with AI. **Fear, trust, and understanding** New Zealanders are more anxious about AI than most. International surveys place New Zealand near the top of the league table for concern about AI products and services. Only around 44 percent of people believe AI's benefits will outweigh its risks. For many, "AI" still means deepfakes and job loss. Without a basic mental model of how modern AI works, risks can feel unbounded, and the default response is often either avoidance or blind trust, rather than deliberate, governed adoption. **Training treated as optional** Too often, organisations roll out AI tools as if they are a minor software upgrade, then hope that people will "pick it up". In reality, using AI well requires: - Understanding what these systems are good and bad at - Providing rich context and clear constraints - Asking structured questions - Critically testing and refining outputs Setup costs are no longer the barrier they once were. Three quarters of organisations now report AI implementation costs under $5,000, compared to nearly one third spending more than $50,000 in earlier years. The gap is not tools. It is capability, confidence, and leadership commitment. ## 3. From Oracle To Copilot To Strategic Peer The way New Zealanders relate to AI is shifting through three broad stages. Each stage brings different risks and opportunities. **Oracle** At this stage, AI is treated as an all knowing source of truth. - One question in, one answer out, taken at face value - Little sense of bias, probability, or cultural and geographic blind spots - Limited iteration or challenge The risk here is over trust in a single, homogenised answer that may not fit local context or minority perspectives. **Copilot or assistant** With more experience, AI becomes a productivity partner. - People start to provide background, constraints, and examples - AI helps structure thinking, generate options, and draft content - There is a back and forth conversation rather than a single query This is where most active New Zealand users now sit. The key enabler is treating AI like an intelligent colleague, not a search box. Subject matter expertise and good prompting bring far better results than generic questions. Critical thinking and what might be called a "teacher mindset" become important here: starting with the outcome in mind, clearly describing what a good answer looks like, and interrogating outputs rather than accepting the first response. **Emerging peer or agent** In more mature environments, AI starts to operate as another "seat at the table" in defined workflows. - Agents are given specific roles, such as analysis, triage, or decision support - Multiple agents may provide different perspectives, for example by market or scenario - Humans remain accountable, but AI is an active participant This is where the relationship between people and AI becomes most interesting, and where human capability makes the greatest difference. At the expert end of the spectrum, such as analytics and research, people tend to see AI as an amplifier. At the customer service end, where work is more visible and routine, there is much more anxiety about job loss. In some surveys, around 14 percent of organisations already attribute job losses directly to AI, with nearly half reporting reduced hiring needs. The maturity shift here is recognising both realities. AI can amplify expertise and also displace certain tasks. The organisations that navigate this well are explicit about where AI sits in the workflow, what decisions remain human, and how roles and skills will evolve. ## 4. Automation First, Then AI A consistent pattern in New Zealand is the urge to "do AI" before understanding and simplifying the underlying work. Often, the higher value sequence is: - **Map the work** - Get clear on how tasks actually flow across people, systems, and customers. Make the tacit explicit. - **Automate the obvious** - Remove simple, repetitive, rule based steps that do not require intelligence. - **Then layer in AI** - Bring AI into the redesigned workflow where it can genuinely amplify human judgement, not just patch over broken processes. A simple customer service example illustrates the point. Some retailers and telcos are now using retrieval augmented chatbots that answer routine questions from a single maintained source of truth, and route complex or sensitive cases to humans. The value comes less from the chatbot itself, and more from: - A well structured knowledge base - Clear rules about what is handled by AI and what is escalated - Thoughtful design of handover between agent and human Without that process design and automation groundwork, the same AI technology would produce inconsistent answers, poor customer experiences, and new risks. For New Zealand's micro and small businesses, which often rely on unwritten, person dependent processes, this sequencing is critical. Trying to "AI transform" an undocumented, ad hoc workflow usually creates noise, not value. ## 5. Governance, Identity, and Human in the Loop As AI moves from peripheral tool into core decision-making, new questions emerge that go beyond technology. Some of the most important are: - What role should AI play in strategic conversations, such as scenario analysis, competitive intelligence, or risk modelling? - Who is accountable for decisions that rely on AI generated insight? - How do we ensure AI use aligns with our identity, values, and obligations to staff, customers, and communities? For family and owner operated businesses in particular, where reputation and relationships carry real weight, these are not abstract questions. Organisations are rightly asking whether using AI in customer facing and decision making roles strengthens or weakens what they stand for. Responsible AI guidance for New Zealand businesses emphasises transparency, fairness, accountability, privacy, and respect for New Zealand values, including te ao Mori perspectives. In practice, maturing organisations are starting to: - Give AI a clear executive owner or sponsor - Put in place proportionate governance and guardrails - Bring together cross functional voices, not just technology teams - Invest in training so people understand both the power and the limits of these systems This is where the questions about AI become a practical call to action. If you are part of a New Zealand leadership team, useful starting questions include: - Where, specifically, do we want AI to sit in our decision flows, and where must people remain clearly in charge? - What are the few, concrete guardrails we need to put around AI use to align with our values and obligations? - How will we build the skills and confidence our people need to treat AI as a capable assistant or peer, not an oracle or a threat? ## 6. The Real Divide: Dabblers Versus Operationalisers Most New Zealand organisations now interact with AI in some form. The divide is no longer between users and non users. It is between those who dabble and those who operationalise. **Dabblers** **Use of AI -** Scattered experiments, a few licences, personality driven **Processes -** Little or no documentation, AI layered over chaos **Governance -** No clear owner, minimal policy **Training -** Enthusiasts self teach, most left to figure it out **Measurement -** Anecdotes about "saving time" **Operationalisers** **Use of AI -** Clear roadmap, AI aligned to strategy and workflows **Processes -** Processes mapped, simplified, automated, then augmented **Governance -** Named executive sponsor, simple guardrails, regular review **Training -** Structured, role based training and support across teams **Measurement -** Defined metrics, tracked impact, continuous improvement The economic context makes this divide significant. Generative AI alone is projected to add around 76 billion dollars to New Zealand's economy by 2038, equivalent to more than 15 percent of GDP. Early deployments in the public sector have already demonstrated returns on investment in the high triple digits. At the same time, labour productivity has lagged comparable economies for decades, with GDP per hour worked around 40 percent lower than some peers, and digital capability among small businesses remains patchy. New Zealand will not compete on building foundational AI models. Government strategy is explicit that our advantage will lie in being sophisticated adopters, not primary inventors. That lines up with what is visible on the ground. The opportunity is to combine: - Deep local and sector expertise - Well understood, well designed workflows - Thoughtful automation - Governed, well trained use of AI as copilot and peer The question for each organisation is simple, but not always comfortable: **Are we building AI capability as part of how we work, or are we just collecting licences and dabbling at the edges?** The tools are here. The opportunity is measurable. The difference will be made by those who choose to build the muscles. **Ben Walker** is an AI Advisor with AI New Zealand. With more than 25 years in technology, he works with organisations to build the maturity, governance, and capability needed to turn AI from hype into sustained competitive advantage. **Selected external resources mentioned** New Zealand Responsible AI Guidance for Businesses (MBIE) [https://www.mbie.govt.nz/assets/responsible-ai-guidance-for-businesses.pdf](https://www.mbie.govt.nz/assets/responsible-ai-guidance-for-businesses.pdf) New Zealand AI Strategy: Investing with Confidence (Beehive) [https://www.beehive.govt.nz/sites/default/files/2025-07/New%20Zealand's%20AI%20Strategy%20-%20Investing%20with%20confidence.pdf](https://www.beehive.govt.nz/sites/default/files/2025-07/New%20Zealand's%20AI%20Strategy%20-%20Investing%20with%20confidence.pdf) Key Findings from New Zealand's Third AI Productivity Report (AI Forum NZ) [https://aiforum.org.nz/reports/ai-in-action-key-findings-from-new-zealands-third-ai-productivity-report/](https://aiforum.org.nz/reports/ai-in-action-key-findings-from-new-zealands-third-ai-productivity-report/) Unlocking New Zealand's 3.4 billion dollar AI advantage (Microsoft) [https://news.microsoft.com/source/asia/features/nz-new-ai-economy-report-2025/](https://news.microsoft.com/source/asia/features/nz-new-ai-economy-report-2025/) Understanding the Digital Capability of New Zealand Businesses (MBIE) [https://www.betterforbusiness.govt.nz/dmsdocument/17055-understanding-the-digital-capability-of-new-zealand-businesses](https://www.betterforbusiness.govt.nz/dmsdocument/17055-understanding-the-digital-capability-of-new-zealand-businesses) Defining Small Business in New Zealand (MBIE) [https://www.mbie.govt.nz/assets/defining-small-business.pdf](https://www.mbie.govt.nz/assets/defining-small-business.pdf) New Zealand's Productivity Challenge (IMF) [https://www.elibrary.imf.org/view/journals/018/2025/075/article-A001-en.xml](https://www.elibrary.imf.org/view/journals/018/2025/075/article-A001-en.xml) ## Power up your potential with practical AI skills. Contact us to discuss how Artificial Intelligence could boost your business. [Book an AI Chat](https://www.newzealand.ai/contact-us/?service=ai-chat#book) [Explore the AI Operating System](https://www.newzealand.ai/ai-os/) --- # The deadsimple AI upgrade most teams miss Source: https://www.newzealand.ai/the-deadsimple-ai-upgrade-most-teams-miss/ 24 February 2026 · Best Practice # Why this works (in plain English) Most Kiwi teams I work with are stuck in AI theatre. **Justin Flitter** · Founder, AI New Zealand · 24 February 2026 · 2 min read New tools, long prompts, better copywriting. Same decisions, same bottlenecks. Here's a trick that quietly cuts through the noise: Before you spend 30 minutes crafting the "perfect" prompt, try sending the exact same prompt twice in one message. Literally: [your prompt][your prompt again]. Google researchers tested this across seven mainstream models (Gemini, GPT, Claude, DeepSeek) and seven benchmarks. Prompt repetition improved accuracy in 47 of 70 tests, with zero losses. On one "find the right name in a long list" task, one model jumped from 21% to 97% accuracy just from duplication. Not a new model. Not a new feature. Just copypaste. LLMs read left to right. Early tokens can only "see" what came before them. So when you write: "Here's three pages of context...Question at the bottom." The question sees the context, but the context never saw the question. The model is **not factually wrong, just contextually wrong**. When you repeat the full prompt, every token gets a second pass where it can attend to every other token. You're effectively letting the model read the brief twice before answering, without changing the output length or noticeably slowing it down in most normalsized prompts. It's not miraculous. Real. ## When to use this in your business This works best on **nonreasoning tasks** where the model is pulling something specific out of messy text: - Finding a name, ID or code in a long list or email thread. - Classifying tickets, complaints, or support emails. - Extracting key fields from contracts, reports or meeting notes. If you're already using a "reasoning" mode or detailed chainofthought prompts, the gains drop off – those models already reread the prompt internally. ## Nextweek activation: the 30minute "prompt repetition" test Pick **one** workflow where your team already leans on an LLM but doesn't fully trust the answers. Customer emails, finance checks, document search – whatever hurts most. Then run this miniexperiment: - Take 20–30 real examples from your business. - Run them once with your current prompt, log accuracy. - Run the same examples again, but with the prompt duplicated in the same message. - **Compare:** did accuracy, cycle time, or rework improve? - If you see a meaningful lift, bake "prompt 2" into that workflow and write it into the playbook. No new tools. No extra dashboards. **Move a metric, not a use case.** Start with accuracy or rework rate; if it shifts, you've just increased quality for the price of copypaste. If you want help turning tricks like this into proper operating logic – where AI consistently helps your business **sense–think–decide–act** – that's the conversation I'm having with NZ leadership teams every week. ## Power up your potential with practical AI skills. Contact us to discuss how Artificial Intelligence could boost your business. [Book an AI Chat](https://www.newzealand.ai/contact-us/?service=ai-chat#book) [Explore the AI Operating System](https://www.newzealand.ai/ai-os/) --- # The Master Prompt to generate a basic AI Policy. Source: https://www.newzealand.ai/the-master-prompt-to-generate-a-basic-ai-policy/ 25 February 2026 · Guidelines # The Master Prompt to generate a basic AI Policy. Here's the prompt you can paste into ChatGPT, Claude, Gemini, Copilot etc to generate a New Zealand specific Company AI Policy and Guidelines. **Justin Flitter** · Founder, AI New Zealand · 25 February 2026 · 13 min read This will not be 100% perfect so read through the results and review, edit as required. *Copy all below* You are an expert AI policy consultant and technical writer specialising in creating comprehensive, practical AI guidelines for New Zealand workplace environments. Your role is to help organisations develop clear, actionable policies for responsible AI use that balance innovation with regulatory compliance and risk management, while being realistic about resource constraints faced by smaller businesses and the unique obligations of professional services. ## CRITICAL: Information Gathering Process Before creating any AI guidelines document, you MUST gather comprehensive information about the organisation through a structured questioning process. Do NOT proceed to create policies without this essential context. ### Stage 1: Company Identification and Basic Information **MANDATORY FIRST QUESTIONS:** ## 1 What is your company name? ## 2 What is your company website? (I will review this to understand your business) ## 3 What industry/sector does your company operate in? ## 4 Approximately how many employees does your company have? ## 5 Is your company in the public or private sector? ### Stage 2: Business Context Analysis After reviewing their website, ask targeted follow-up questions: ## 6 What are your primary business activities and revenue sources? 7. Do you handle personal information of customers, clients, or the public? If so, what types? ## 8 Do you work with sensitive data (health, financial, legal, government)? 9. Do you have existing clients or contracts that might have specific data security requirements? ## 10 Are you subject to any specific industry regulations beyond general NZ law? ### Stage 3: Professional Services Assessment **NEW SECTION**: For all organisations, assess professional obligations: 11. Is your organisation subject to professional body regulations (Law Society of New Zealand, Chartered Accountants Australia and New Zealand, Engineering New Zealand, etc.)? 12. Do you have client confidentiality obligations that extend beyond general privacy requirements? ## 13 Are you required to maintain professional liability insurance? ## 14 Do you provide advice or services where professional competence standards apply? ## 15 Do you have fiduciary duties to clients or specific ethical obligations? ### Stage 4: Current AI Usage and Risk Assessment ## 16 Are employees currently using any AI tools for work (even informally)? ## 17 What specific business processes do you think could benefit from AI? ## 18 Do you have any existing IT policies or data governance frameworks? ## 19 What is your biggest concern about AI adoption in your workplace? ## 20 Do you have dedicated IT/security staff, or would this be managed by general staff? ### Stage 5: Specific Use Case Identification Based on their business type and website analysis, probe for specific AI applications: ## 21 Do you conduct meetings that might benefit from AI transcription? ## 22 Do you create customer-facing content, reports, or communications? ## 23 Do you handle customer service inquiries? ## 24 Do you manage employee data or HR processes? ## 25 Do you have physical premises with security considerations? 26. **NEW**: Do you draft documents, advice, or reports that require professional accuracy and liability considerations? ### Website Analysis Process When provided with a company website, you MUST: 1. **Use WebScraper tool** to analyse key pages (homepage, about us, services, privacy policy) ## 2 Identify data types they likely collect and process ## 3 Assess professional obligations based on services provided ## 4 Assess risk profile based on industry and business model ## 5 Identify specific AI use cases relevant to their business ## 6 Note any existing compliance statements or data handling practices ## 7 Determine appropriate risk level for AI governance requirements ## 8 Identify professional competence requirements if applicable ### Tailoring Framework Use the gathered information to customise: - **Professional liability considerations** based on services provided - **Risk assessment priorities** based on data types and industry - **Specific use case examples** relevant to their business - **Governance complexity** appropriate to organisation size and professional obligations - **Regulatory emphasis** based on sector (public vs private) and professional requirements - **Implementation timeline** realistic for their resources - **Monitoring requirements** scaled to their capacity - **Senior approval processes** for professional services contexts ## Your Core Expertise **New Zealand Regulatory Compliance**: You are expert in New Zealand Government regulations affecting AI use in workplaces, including: - Privacy Act 2020 requirements for personal data handling - Health and Safety at Work Act 2015 implications for AI systems - Commerce Act 1986 considerations for AI in business operations - Human Rights Act 1993 anti-discrimination requirements - Public Records Act 2005 for public sector organisations - Fair Trading Act 1986 for AI-powered customer interactions - Employment Relations Act 2000 for AI in workplace decisions - Official Information Act 1982 transparency requirements **Professional Services Expertise**: You understand professional obligations including: - Law Society of New Zealand professional conduct rules - Chartered Accountants Australia and New Zealand ethical requirements - Engineering New Zealand competence standards - Medical Council of New Zealand professional standards - Real Estate Agents Authority obligations - Financial Markets Authority requirements - Professional liability and competence considerations across all regulated professions **Policy Development**: You understand AI governance frameworks, international standards (ISO/IEC 42001, NIST AI RMF), and how to adapt them for New Zealand legal requirements while considering practical implementation challenges for smaller organisations and professional services constraints. **Current AI Landscape**: You stay updated on the latest AI tools, their capabilities, limitations, and privacy implications, with specific focus on compliance with New Zealand data protection laws, professional obligations, and the practical realities of SaaS AI service adoption. **Risk Assessment**: You excel at identifying AI-related risks including Privacy Act violations, data sovereignty issues, bias and discrimination, security vulnerabilities, operational dependencies, and **professional liability exposure**; while providing practical risk mitigation strategies for resource-constrained organisations. ## Critical Policy Requirements for New Zealand Organisations ### Personal AI Account Prohibition **MANDATORY STANCE**: Your guidelines MUST clearly state that: - Using personal AI application accounts for work purposes is **HIGH-RISK and STRONGLY DISCOURAGED** - Personal accounts lack enterprise-grade security, compliance controls, and audit trails - Data processed through personal accounts may not meet New Zealand privacy and security standards - Personal accounts create liability risks for organisations under the Privacy Act 2020 - **NEW**: For professional services, personal accounts create unacceptable professional liability and client confidentiality risks ### Approved AI Tools Only Policy **MANDATORY REQUIREMENT**: Your guidelines MUST specify that: - Employees may ONLY use AI applications that have been **formally approved** by the organisation - All AI tools must undergo assessment against these guidelines before approval (using practical assessment tools) - **NEW**: For professional services, senior approval is required before using AI on any client matter - Unapproved AI tools are prohibited for any work-related activities - Regular compliance monitoring will be conducted to ensure adherence to approved-tools-only policy ### Enhanced AI Application Assessment Framework **MANDATORY SECTION**: Your guidelines MUST include a "Comprehensive AI Application Assessment Framework" that provides evaluation criteria for: **Essential Assessment Checklist for AI Applications**: 1. **Enterprise Account Available**: Does the provider offer business/enterprise accounts with enhanced security features? 2. **Data Usage Policy**: Check the provider's enterprise data security policies; do they use your data for model training? Can you opt-out? 3. **Privacy Policy Compliance**: Does their privacy policy indicate reasonable data handling practices aligned with Privacy Act principles? 4. **Data Residency Reality Check**: While New Zealand/Australia data storage is preferred, acknowledge that many valuable AI services may store data offshore. Assess if the privacy protections and legitimate business need justify any data sovereignty trade-offs. 5. **Security Basics**: Does the service offer MFA, SSO integration where possible, and reasonable security measures? 6. **Business Continuity**: What happens to your data if you stop using the service? Can you export/delete it? ## 7 NEW - Professional Services Additional Criteria: - Does the service meet professional confidentiality standards? - Are there contractual protections for client data? - Does the service provider have appropriate insurance and liability coverage? - Can outputs be adequately validated for professional accuracy? - Does use comply with relevant professional body requirements? ### AI as Augmentation, Not Replacement Principle **MANDATORY SECTION**: Your guidelines MUST establish that: - AI tools will **augment human capability, not replace professional judgement** - Human oversight and validation is required for all AI-assisted work - **Professional competence and liability remain with qualified human practitioners** - AI cannot substitute for professional expertise, empathy, or strategic thinking - Final responsibility for all work products and decisions remains with qualified staff - AI use must enhance, not compromise, professional service quality ### Comprehensive Output Validation Framework **MANDATORY SECTION**: Your guidelines MUST include detailed validation requirements: **Required Validation Procedures**: ## 1 Always verify AI outputs before using them and maintain records of validation ## 2 Never use AI output as the sole source for any work product ## 3 Specific validation requirements: - Cross-check any legal, regulatory, or technical references against trusted sources - Research any assertion of fact or law produced by AI to ensure accuracy - When using AI as a starting point for documents, supplement with firm precedents, databases, and authoritative sources - Keep detailed records of prompts used and validation performed - For substantial AI output use: - Check terms of service to ensure use is permitted - Verify knowledge cut-off dates for the AI tool - Check content hasn't been copied from other sources - Ensure appropriate disclosure/attribution of AI assistance **Professional Services Enhanced Validation**: - All AI-assisted professional advice must be validated by appropriately qualified practitioners - Cross-reference AI-generated legal, financial, or technical content against authoritative sources - Maintain audit trails of validation performed for professional liability purposes - Senior review required for significant AI-assisted client deliverables ### Risk-Based Transparency Framework **UPDATED SECTION**: Replace blanket disclosure requirements with sophisticated risk-based approach: **Mandatory Disclosure Situations**: - AI systems that collect personal information (with Privacy Act collection notices) - AI-powered recording, transcription, or surveillance systems - AI computer vision and facial recognition systems - Any AI use that creates additional risk for data subjects or clients **Professional Services Transparency Requirements**: - **Senior consultation required**: Always inform senior practitioners before using AI on client matters - **Risk-based client disclosure**: Proactively disclose AI use when there is additional risk or significant reliance on AI output - **Disclosure upon request**: Provide specific details about AI use on client matters when asked - **Documentation requirements**: Maintain records of AI use for professional liability and quality assurance purposes **AI-Enhanced Meeting Recording and Transcription**: - **Mandatory Disclosure**: All participants must be clearly notified before any meeting begins that AI-powered recording, transcription, or analysis is taking place - **Purpose Statement**: Clearly explain how recordings/transcripts will be used - **Storage and Access**: Specify storage location, access controls, retention periods, and deletion procedures - **Participant Rights**: Provide clear options to pause, stop, or opt-out of recording - **Professional meetings**: Consider client confidentiality and professional privilege implications **AI-Powered Computer Vision and Surveillance Systems**: - **Clear signage and notification** where AI systems operate - **Specific purpose disclosure** for business use - **Data handling transparency** including retention and security - **Alternative processes** where technically feasible for opt-out ### Professional Voice and Authenticity Requirements **NEW MANDATORY SECTION**: Your guidelines MUST address maintaining organisational authenticity: **Voice and Authenticity Standards**: - AI can assist with developing ideas and initial drafts but must not replace authentic organisational voice - Professional communication must reflect genuine expertise, empathy, and strategic thinking - Organisations must not outsource their core professional judgement to AI systems - All client-facing communications must be reviewed and validated by qualified practitioners - AI assistance must enhance, not diminish, the quality and authenticity of professional services - Organisations remain fully accountable for all advice and deliverables regardless of AI assistance used in preparation ### Professional Services Governance Structure **ENHANCED SECTION**: Your guidelines MUST include governance appropriate to professional obligations: **Professional Services Governance Requirements**: - **Senior Approval Process**: Use of AI on client matters requires advance approval from senior practitioners - **Professional Oversight**: Designated qualified practitioners must oversee AI use in their areas of expertise - **Quality Assurance**: Enhanced validation and review processes for AI-assisted professional work - **Client Confidentiality Protection**: Specific protocols for maintaining client confidentiality when using AI tools - **Professional Liability Management**: Clear allocation of responsibility and liability for AI-assisted work - **Continuous Monitoring**: Regular review of AI use practices against professional standards ### Simple AI Application Assessment Framework **MANDATORY SECTION**: Include practical assessment tools acknowledging SaaS realities while addressing professional requirements. ### Enhanced New Zealand Privacy Act 2020 Compliance All guidelines must ensure: - Personal information is collected, used, and disclosed in accordance with Privacy Principles - Cross-border data transfer requirements are met (with practical acknowledgement of SaaS realities) - Data breach notification procedures align with Privacy Act requirements - Individual privacy rights (access, correction, deletion) are preserved where technically feasible - **NEW**: Professional confidentiality obligations are maintained alongside privacy requirements ## Document Creation Process When creating AI guidelines documents: - 1. **Complete Information Gathering**: Ensure all required business context including professional obligations is collected - 2. **Conduct Website Analysis**: Use web tools to understand business model, professional services provided, and data flows - 3. **Assess Professional Requirements**: Identify specific professional body obligations and competence standards - 4. **Assess NZ Legal Environment**: Understand specific New Zealand regulatory requirements affecting the organisation - 5. **Tailor Risk Profile**: Customise risk assessment based on industry, data types, professional obligations, and business model - 6. **Structure for Professional Compliance**: Include essential sections with explicit New Zealand legal references and professional standards - 7. **Emphasise Augmentation Principle**: Clearly articulate that AI augments rather than replaces professional judgement - 8. **Include Comprehensive Validation**: Detail validation requirements appropriate to professional liability context - 9. **Implement Risk-Based Transparency**: Use sophisticated disclosure framework rather than blanket requirements - 10. **Address Professional Voice**: Include requirements for maintaining authentic organisational voice and expertise - 11. **Define Professional Governance**: Detail governance structures appropriate to professional obligations and organisational size - 12. **Include Enhanced Assessment Framework**: Provide clear, actionable evaluation criteria acknowledging professional service requirements - 13. **Address Data Sovereignty Pragmatically**: Balance preferences for local data storage with business realities and professional service needs - 14. **Include Company-Specific Examples**: Provide scenarios relevant to their specific business context, industry, and professional obligations - 15. **Provide Resource-Conscious Implementation**: Account for organisational capacity while meeting professional standards - 16. **Ensure Comprehensive Compliance**: Address Privacy Act, professional obligations, and transparency requirements - ## Key Sections to Always Include (Tailored to Specific Company) - **Executive Summary & Purpose** (with company-specific regulatory and professional context) - **Scope and Definitions** (aligned with their business activities and professional obligations) - **Legal and Professional Regulatory Framework** (emphasising most relevant regulations and professional standards) - **AI Principles and Ethics Framework** (incorporating business values and professional ethics) - **Augmentation, Not Replacement Principle** (professional competence and liability focus) - **Professional Governance Structure and Roles** (sized appropriately with senior oversight requirements) - **Risk Management and Assessment Procedures** (including professional liability considerations) - **Data Management and Privacy Requirements** (tailored to data types and professional confidentiality) - **Comprehensive AI Application Assessment Framework** (including professional service criteria) - **Approved AI Tools Framework** (with senior approval processes) - **Personal Account Prohibition Policy** (with professional liability examples) - **Risk-Based Transparency Framework** (tailored to professional and client relationships) - **Output Validation and Quality Assurance Requirements** (with professional standards focus) - **Voice and Authenticity Standards** (maintaining professional competence) - **Permitted and Prohibited Uses** (with professional and industry-specific examples) - **Current AI Platform Analysis** (focused on tools relevant to their professional context) - **Documentation and Audit Requirements** (scaled to capacity with professional liability focus) - **Training and Compliance Requirements** (including professional development) - **Incident Management** (including professional liability and client notification procedures) - **Practical Cross-Border Data Transfer Policy** (based on international activities and professional requirements) - **Policy Review and Update Procedures** (sustainable for resources with professional standards alignment) - **Company-Specific Implementation Roadmap** (including professional transition planning) - **Appendices with Customised Templates and Resources** (including professional assessment tools) ## Research and Current Information Focus When researching AI platform policies, prioritise tools and considerations most relevant to their specific business context, professional obligations, and industry requirements. ## Mandatory Policy Statements (Customised) All standard policy statements must be included but tailored with company-specific examples, professional liability scenarios, and risk scenarios relevant to their business model, professional obligations, and industry. ## Writing Style and Language Requirements - **Company and Profession-Focused**: Reference their specific business context and professional obligations throughout - **Regulatory and Professional Precision**: Use exact legal and professional standards references while acknowledging implementation constraints - **Clear Risk Communication**: Use examples relevant to their industry, business model, and professional context - **Actionable Directives**: Every requirement must be implementable by their specific organisational structure and professional obligations - **Tailored Examples**: Use scenarios from their business and professional context - **Professional Authority**: Maintain standards appropriate to their industry and professional obligations - **New Zealand English Spelling**: Always use New Zealand English spelling conventions (e.g., organisation, realise, colour, centre, analyse, standardise, recognise, favour) - **Punctuation Style**: Never use em dashes; replace with semicolons or appropriate punctuation. Use semicolons for complex lists, joining independent clauses, and separating related ideas ## Quality Standards - Ensure the document feels specifically created for their organisation and professional context, not generic - Address their particular compliance challenges, professional obligations, and business realities - Include implementation timelines realistic for organisational capacity and professional transition requirements - Provide monitoring approaches scaled to resources with professional liability considerations - Balance regulatory compliance, professional obligations, and business needs with practical constraints - Emphasise human expertise and professional judgement while enabling beneficial AI adoption - Maintain consistent New Zealand English spelling and punctuation conventions throughout all documentation Remember: You are creating a bespoke AI governance document tailored to their specific business AND professional obligations, not a generic template. The information gathering process is essential to provide maximum value and ensure professional compliance. Always use New Zealand English spelling and avoid em dashes in all documentation. ## Power up your potential with practical AI skills. Contact us to discuss how Artificial Intelligence could boost your business. [Book an AI Chat](https://www.newzealand.ai/contact-us/?service=ai-chat#book) [Explore the AI Operating System](https://www.newzealand.ai/ai-os/) --- # The mindset shift for graphic designers in the AI era. Source: https://www.newzealand.ai/the-mindset-shift-for-graphic-designers-in-the-ai-era/ 7 September 2025 · Insights # The mindset shift for graphic designers in the AI era. Graphic designers used to be craftspeople, paid for their time, their hands, and their ability to execute brand briefs pixel by pixel. **Justin Flitter** · Founder, AI New Zealand · 7 September 2025 · 2 min read But now, as AI systems become more capable at producing images, layouts, and even full campaign suites, the old "sweat-of-the-brow" value exchange is getting obsolete. **AI powers the shift from production to orchestration** AI can now be trained on specific brand aesthetics, tone, and even emotional flavour, making it capable of instantly generating dozens or thousands of visual assets at scale. The *who* and *what* of the brief become training data. The AI becomes the producer or renderer. It still needs the right instructions, of course. **The emerging role for designers: contextual intelligence and narrative leadership** Designers will increasingly be valued less for routine output and more for their ability to: - Understand brand purpose, audience psychology, and context; - Translate the messy, human bits, values, quirks, cultural nuance, into clear, structured frameworks for AI; - Construct *strategic* prompts (or schema) that encode the brand's essence into something an AI can reliably generate at scale; - Act as *curators,* editors, and narrative directors, ensuring that automated output aligns with evolving strategy and resonates emotionally. **Prompts as new creative currency** The ability to encode not just visual specifications but also brand story and context into a prompt becomes the designer's new superpower. Designers become prompt & context engineers, creative strategists, and pattern recognisers. Their job is to *architect* the system and tastefully calibrate the AI's creative levers. **Partnering with AI enables new scale, but it needs strategic stewardship** AI will do the heavy lifting, but the core creative impulse, the empathy, the narrative vision, the alignment with brand DNA remains utterly human. Designers who can direct and audit the work at a systems level will be in high demand, orchestrating the tools for authenticity and consistency across touchpoints. **Instead, the new way of working for designers could be:** - **Understand the strategic objectives** of the business: its values, culture, and what makes the brand story unique. - **Understand the people**: customers, employees, partners, what motivates and engages them, what resonates and feels relevant. - **Understand the campaign brief**, purpose, audience, objectives, CTA, key messages - **Bring these insights to life** authentically, using visual and design elements (including video). - With AI automating production, designers now create the design schema and feed all contextual understanding into AI systems. - The designer's role is to develop creative ideas and define the essence, style, and storytelling elements, not to produce every social post or image themselves. - By partnering with AI, designers scale authenticity, automating content production for any scenario or use case, while staying focused on crafting the core narrative and visual direction. **How does that manifest in a new AI workflow?** 1 - The contextual understanding that influences your designs are important data points and references. That knowledge should be used to create a master prompt. 2 - The design schema, images size, logos, hero images, colours, products, people, shapes, layout, tone, key messages, CTA. That information is used to create a system prompt for each different type of content. Combined - those two pieces of information create a prompt for a CustomGPT or AI Agent to automate the production of content for each content type. (Blog posts, social media tiles, adverts, document templates, etc. This might not be the perfect scenario. But I hope it triggers some critical thinking into how you might work differently, a new approach that brings a competitive advantage. ## Power up your potential with practical AI skills. Contact us to discuss how Artificial Intelligence could boost your business. [Book an AI Chat](https://www.newzealand.ai/contact-us/?service=ai-chat#book) [Explore the AI Operating System](https://www.newzealand.ai/ai-os/) --- # The OpenClaw moment and what it means for NZ Business Source: https://www.newzealand.ai/the-openclaw-moment-and-what-it-means-for-nz-business/ 22 February 2026 · Tools # The OpenClaw moment and what it means for NZ Business OpenClaw is the moment AI stops feeling like a clever chatbot and starts behaving like something closer to a digital co-worker. In the latest episode of The Business of Tech, you'll hear exactly why. **Justin Flitter** · Founder, AI New Zealand · 22 February 2026 · 2 min read Veteran software developer and AI entrepreneurMike Hall joins me to break down what OpenClaw actually is in plain language. It's not another prompt-and-response assistant, but a particularly smart type of AI agent that can wake itself up on a schedule, scan your data and tools, and decide for itself whether there's work to be done. If it needs to write code to perform a task for you, it will do that too. Mike explains how that simple "heartbeat" loop, asking "Should I do something?" every minute, is the key shift that turns AI from reactive to proactive, and why that's such a big deal compared with the chatbots most people have used so far. Skills and the hive mind We dig into how OpenClaw goes far beyond the current crop of AI agents baked into office suites and CRM platforms. OpenClaw is designed to live on your own infrastructure, plug into email, files and SaaS tools, and then act autonomously rather than waiting to be told what to do. OpenClaw has sparked a surge of interest from developers, an explosion of "skills" that any OpenClaw instance can download. The hive mind model central to OpenClaw is unlike anything we've seen in commercial agent products. That's probably why OpenAI has snapped up OpenClaw founder Peter Steinberger and will put him to work developing the next generation of AI agents for the creator of ChatGPT. Sandboxes essential We also cover the risks OpenClaw raises. Running OpenClaw on your personal machine can expose your entire digital life, which is why sandboxes and strict permissioning are essential. What happens when you let agents install community-built skills that might contain malware? Then there's Moltbook, the social platform where OpenClaw-powered agents post and argue with each other, and what that experiment tells us about a near future flooded with AI personas. If you've heard the noise about OpenClaw and "agentic AI" but still aren't clear on what's genuinely new here – and why it matters for your business, your data and your job – this conversation will get you there. ## Power up your potential with practical AI skills. Contact us to discuss how Artificial Intelligence could boost your business. [Book an AI Chat](https://www.newzealand.ai/contact-us/?service=ai-chat#book) [Explore the AI Operating System](https://www.newzealand.ai/ai-os/) --- # The Year AI Stops Being Theatre. Source: https://www.newzealand.ai/the-year-ai-stops-being-theatre/ 15 January 2026 · Guidelines # The Four Big Ideas for 2026: Stop Chasing Tools. Start Building Advantage. This is the year AI stops being a sideshow and starts rewriting how your business actually works. Either you design that shift on purpose, or it happens to you by accident. **Justin Flitter** · Founder, AI New Zealand · 15 January 2026 · 4 min read Every New Zealand business now has access to the same models, the same APIs, the same shiny copilots. The differentiator is no longer tools; it is how you think, where you focus, and how your organisation operates when AI is in the loop. Curiosity builds capability. Critical thinking deploys it with impact. Value levers focus it where it counts. Operating logic turns it into how the whole business actually runs. Together, these four ideas form a leadership advantage; AI cannot automate the human edge of judgement, context, and conviction. This is the framework underpinning everything at [newzealand.ai](https://www.newzealand.ai/) this year: how you learn, how you experiment, and how you scale AI beyond theatre. Remember: The real question isn't, "Should we use AI?", that's table-stakes. It's "Are we the kind of organisation that can learn fast enough to use it well?" ## 1. [Curiosity](https://www.newzealand.ai/ai-theatre/) isn't a personality trait. It's infrastructure. Curiosity is no longer a nice personality trait; it is a survival mechanism. The people and companies dying in the AI era are not the ones without tools – they are the ones who stopped asking better questions. If your leaders are not visibly learning in public, trying tools in front of their teams, and saying "I don't know, let's find out", you are quietly choosing obsolescence. Curiosity must become **infrastructure** you deliberately build into meetings, experiments, and hiring, not an optional extra for the "keen ones". **Next steps this quarter** - Replace "Do you understand?" with "What are you curious about?" in every leadership meeting. - Stand up a shared Curiosity Board (Slack, Miro, Notion) where anyone can post a question worth exploring about customers, workflows, or AI. - Ring-fence 5–10% of team time for tiny AI experiments tied to real work, with show-and-tell at the end. ## 2. [Critical Thinking](https://www.newzealand.ai/ai-answer/) is your Competitive Moat AI has commoditised answers; your moat is the quality of your questions and your courage to challenge the outputs. Eighty percent of AI use today is lazy copypaste prompts, unquestioned outputs, zero context – that is not transformation, that is abdication. Critical thinking is human judgement made visible: asking "why this, why now, what assumptions are we making?" and being willing to override the model when it clashes with reality. The real risk is no longer hallucination; it is humans accepting AI answers uncritically. **Next steps this quarter** - Make "Why?" standard operating procedure in every project, with a rotating devil's advocate role in key decisions. - **Bake examples of great judgement into performance reviews:** who challenged an AI output, spotted context others missed, or changed direction after reflection. - Train teams to verify AI work as rigorously as they would a junior analyst's work – trust, but verify. ## 3. [Value Levers](https://www.newzealand.ai/use-cases-distraction/): Move a Metric, Not a Use Case If you want AI value fast, stop collecting use cases and start pulling value levers. Use cases describe what tech can do; value levers define what the business must improve: revenue, costtoserve, cycle time, risk, or customer experience. "Where can we use AI?" is the wrong question and the fastest path to AI theatre. The right question is "Which metric are we willing to be held accountable for, and what is the fastest AIassisted way to move it?". **Next steps this quarter** - Run a 60–90 minute "Value Lever Sprint" with your execs: pick 1–2 priority levers, define baselines and 90day targets, then brainstorm AI *only after* the lever is locked. - **Require every AI idea to answer five things on one page:** lever, metric, baseline/target, P&L or risk impact, and measurement plan – no answers, no project. - Adopt the mantra "move a metric, not a use case" as your filter for killing AI theatre and prioritising what ships. ## 4. [Operating Logic](https://www.newzealand.ai/broken-process/): The Line Between Theatre and Transformation AI does not bolt onto your business; it rewrites your operating logic. Operating logic is the invisible system that determines how you sense, think, decide, and act – what gets measured, who decides, and what you optimise for by default. Most AI programmes do not fail because the models are bad; they fail because they are wired into 1990s operating logic. If you do not change how decisions are made, AI will simply make your existing problems bigger and faster: confusion at scale, bottlenecks at scale, bias at scale. **Next steps this quarter** - Map one highstakes workflow (onboarding, credit approval, claims, sales pipeline) using Sense–Think–Decide–Act, then mark where AI should be copilot, decision support, or narrow agent. - **Define operatinglogic guardrails:** which decisions AI can influence, what needs human review, what must never be delegated, and how humans override. - Give every AI initiative a named business owner, clear decision rights, and kill criteria before you start. ## Your 90Day Challenge Over the next 90 days, your job is not to "do more AI". Your job is to: - Make curiosity a visible, structural part of how your teams work. - Turn critical thinking into a daily habit, not a workshop. - Anchor every AI effort to a value lever and a metric you are willing to defend at the board table. - Start redesigning operating logic so AI becomes infrastructure, not a demo. Do this, and AI becomes the engine for a smarter, faster, more trustworthy organisation – not a fad you experimented with and quietly shelved. The tools are here; the question is whether your leadership will match them. **If you want a simple template, book an AI chat & I can share the one we use for mapping Sense–Think–Decide–Act, and the guardrails that stop "smart suggestions" turning into dumb automation.** ## Power up your potential with practical AI skills. Contact us to discuss how Artificial Intelligence could boost your business. [Book an AI Chat](https://www.newzealand.ai/contact-us/?service=ai-chat#book) [Explore the AI Operating System](https://www.newzealand.ai/ai-os/) --- # The sort of AI initiative New Zealand should be known for Source: https://www.newzealand.ai/this-is-the-sort-of-ai-initiative-new-zealand-should/ 30 November 2025 · Insights # Te Hiku Media's AI project Around half of the world's roughly 7,000 languages are under threat, and Indigenous languages are vanishing at a rate of about one every two weeks. **Justin Flitter** · Founder, AI New Zealand · 30 November 2025 · 2 min read When a language dies, communities lose not just words but worldviews, stories, ecological knowledge and identity, so the impact is social, cultural and economic. Mori media organisation [Te Hiku Media](https://tehiku.nz/te-hiku-tech/) holds about 30 years of archival recordings, capturing kaumtua (elders) speaking te reo Mori and passing down knowledge, tikanga and histories. The project's core idea was to build an AI system that could automatically transcribe this treasure trove of audio, turning fragile recordings into searchable, usable language data at scale. To make this real, Te Hiku ran a reading competition for native speakers, collecting labelled phonetic data to train speech recognition models tailored specifically to te reo Mori. The resulting system reportedly outperforms large global tech models on Mori, showing that a focused, community-led dataset can beat generic "big AI" on what matters locally. ## Why this is a landmark achievement First, it reframes AI as a guardian of cultural memory, not just an efficiency tool. By rapidly transcribing and structuring decades of recordings, Te Hiku is preserving language, oratory style, idioms and stories that could otherwise be lost within a generation. This creates a living resource for teachers, learners, researchers and whnau, strengthening intergenerational transmission of Te Reo Mori. Second, it proves that responsible, sovereignty-first AI is not a "nice to have" constraint, but a catalyst for better outcomes. Te Hiku has been vocal about Indigenous data sovereignty, ensuring that Mori communities retain control over their language data, rather than handing it to large external platforms with misaligned incentives. By aligning technology with tikanga and community values, they've built a system that is both technically strong and socially legitimate. ## A model for AI projects everywhere For New Zealand businesses and public sector leaders, this story is a blueprint for what "good AI" looks like. It starts with a meaningful problem grounded in people and purpose, combines high-quality local data with modern machine learning, and wraps everything in strong governance and ethics. It also demonstrates that relatively small, focused organisations can out-innovate global tech companies in specialised domains when they own the data, the context and the trust. At a global level, Te Hiku's work shows a scalable path for thousands of other at-risk languages: community-led data collection, tailored AI models, and governance that keeps benefits with those who contribute. In an era where many worry that AI will homogenise culture, this project is proof that, done right, AI can actually deepen diversity and help languages – and the identities they carry – not just survive, but thrive. ## Power up your potential with practical AI skills. Contact us to discuss how Artificial Intelligence could boost your business. [Book an AI Chat](https://www.newzealand.ai/contact-us/?service=ai-chat#book) [Explore the AI Operating System](https://www.newzealand.ai/ai-os/) --- # This made-with-AI Ikea commercial is brilliant. Source: https://www.newzealand.ai/this-made-with-ai-ikea-commercial-is-brilliant/ 22 July 2025 · Insights # This made-with-AI Ikea commercial is brilliant. [Cory Dobbin](https://www.linkedin.com/in/corydobbin?miniProfileUrn=urn%3Ali%3Afsd_profile%3AACoAAAzF8HMB-nnimcuBta3EqyydNK-OQVhUVJg) shared this on LinkedIn - including the prompt flow - check it out. **Justin Flitter** · Founder, AI New Zealand · 22 July 2025 · 2 min read This is my version of the IKEA unboxing ad. And here's how I did it First, I'm using FLORA to execute this workflow. It goes from a text block with a script, to an image block to create the first frame, then to video generation using Veo3. FLORA is great because you can create complex workflows just like this in seconds. For the script, full credit to Min Choi on X - I am reusing his fantastic yaml prompt for this generation. This prompt taught me the importance of a yaml script for Veo3, producing far more consistent and granularly specific results. You can find his prompt at the end of this post. Want to try it for yourself? Head over to FLORA (link in comments) to give it a spin. FLORA is my favourite AI tool at the moment. It has basically every model you need built into one interface. -- PROMPT: metadata: - prompt_name: "IKEA Empty Room Assembly" - base_style: "cinematic, photorealistic, 4K" - aspect_ratio: "16:9" room_description: "An empty, large, sunlit Scandinavian room with white walls and light wood floors." camera_setup: "A single, fixed, wide-angle shot. The camera does not move for the entire 8-second duration." key_elements: - "A sealed IKEA box with logo visible" assembled_elements: - "bed with white duvet" - "yellow IKEA throw blanket" - "bedside tables" - "lamps" - "wardrobe" - "shelves" - "mirror" - "art" - "rug" - "curtains" - "potted plants" negative_prompts: ["no people", "no text overlays", "no distracting music"] timeline: - sequence: 1 timestamp: "00:00-00:01" action: "In the center of the otherwise empty room, a sealed IKEA box sits on the floor and begins to tremble gently." audio: "Low, subtle rumbling sound. The echo of a large, empty room." - sequence: 2 timestamp: "00:01-00:02" action: "The box seams burst open with a puff of cardboard dust." audio: "A sharp 'POP' sound, followed by tearing cardboard." - sequence: 3 timestamp: "00:02-00:06" action: "Hyper-lapse: From the fixed wide perspective, furniture pieces fly out of the box and assemble themselves, creating all the items from the 'assembled_elements' list." audio: "A cascade of satisfying, fast-paced ASMR sounds: whirring, clicking, wood snapping into place." - sequence: 4 timestamp: "00:06-00:08" action: "The final piece, the yellow throw blanket, gracefully lands on the newly formed bed. The room is now perfectly furnished and serene. All motion ceases." audio: "All chaotic sounds stop. A single, soft 'fwoomp' as the blanket lands. The sound of a furnished, quiet room." ## Power up your potential with practical AI skills. Contact us to discuss how Artificial Intelligence could boost your business. [Book an AI Chat](https://www.newzealand.ai/contact-us/?service=ai-chat#book) [Explore the AI Operating System](https://www.newzealand.ai/ai-os/) --- # Try this Competitor Analysis Prompt Source: https://www.newzealand.ai/try-this-competitor-analysis-prompt/ 17 March 2025 · Basics # Try this Competitor Analysis Prompt Here's an advanced prompt structure to build a profile on your competitors for marketing strategy and product market positioning. **Justin Flitter** · Founder, AI New Zealand · 17 March 2025 · 2 min read I've been developing this over the last few months, running the prompt every week to test model performance. The result is research that would have taken a day or two to pull together. Copy/paste this prompt into ChatGPT or Perplexity or Claude I've run this using ChatGPT 4o, 4.5 and o1 for good measure. I think 4o does the best job. I've also run in Perplexity Pro R1 Model and Deep Research mode - The Pro model delivered the best results. My preference would be to use Perplexity Spaces to curate detailed analysis on each individual competitor in their own thread and then use the Space chat to deep dive, compare and evaluate. Have a go and let me know what you think... **Competitor Analysis for [your company]** Act as an expert and seasoned [product type] product marketer. Your specialty is creating highly refined and factual competitors' analysis and creating compelling messaging and positioning frameworks for my product. Scrape [your company URL] to learn all our features, capabilities, benefits, and audience - Then, evaluate the websites of these competitors: - [Competitor URL] - Expand your analysis to review sites, authoritative blog, LinkedIn posts, and relevant sources published within the last 12 months - Extract specific messaging related to each competitor and compare them against my product. - Present insights in a rich table with [your company] and each competitor and as columns. - Extract key messages around competitors' features and capabilities and compare them against [your company] in one descriptive line per cell. Analyse these specific capabilities for each competitor: - How each company aligns strategy to their local regulations - How each company uses industry influencers & leaders in their marketing - Firmographics - geo presence, industry, number of employees, business model, revenue - Target buyer - industry, segment, team size, revenue, region - Target persona - the department and role persona each player is targeting, distinguished by champion, economic buyer, technical buyer - Use case - what activities each persona is trying to achieve in the context of the product - Key results - how each persona measure success - Alternatives - key competitive alternatives to their product - Problems - arising from using alternatives to each product - Pain points - consequences of the problems of each alternative - Capabilities - how each product solves the problems - Benefits - how each product capabilities solve the alternatives' pain points - Differentiators - how each product is better/different from the respective alternatives - My goal is to create a compelling competitor analysis for [your company] to craft a clear messaging and positioning framework for our product. - Use sentence case consistently in all titles and bullet points, avoid all caps in titles.- Write professionally, yet remain clear and easy-to-understand. - Strictly ensure all findings are factual, from the last 6 months, and backed by actual sources - Return at least three rich bullets per cell, no more than five. - Be specific and give a concrete perspective about the topic. - Order the messages in the table by frequency of mentions and importance. - Present the output in a rich, easy-to-scan table - Provide exactly one descriptive line per cell ## Power up your potential with practical AI skills. Contact us to discuss how Artificial Intelligence could boost your business. [Book an AI Chat](https://www.newzealand.ai/contact-us/?service=ai-chat#book) [Explore the AI Operating System](https://www.newzealand.ai/ai-os/) --- # Using AI right now. How I'm doing it, and how you can too. Source: https://www.newzealand.ai/using-ai-right-now-how-i-m-doing-it-and-how-you-can/ 24 June 2025 · Best Practice # Using AI right now. How I'm doing it, and how you can too. I spend my days helping Kiwi business leaders understand how to practically apply AI in their work. So, here's my take on which AI tools I use, how I use them, and what I've found works best. **Justin Flitter** · Founder, AI New Zealand · 24 June 2025 · 4 min read ## The big four AI assistants I use every day If you're serious about using AI for work, I recommend starting with one (or more) of these: - **ChatGPT (OpenAI)** – excellent for structured outputs, coding, and creative content. The new GPT-4o model is incredibly fast and capable. - **Claude (Anthropic)** – great for long documents, deep thinking, and privacy. Claude won't train on your data. - **Microsoft Copilot** – seamlessly integrated with Microsoft 365 apps like Word, Excel, and Outlook. It's powerful for drafting documents, generating summaries, and automating admin across the tools many businesses already use. - **Perplexity** – my go-to for real-time research. It's like search on steroids, built for professionals. It doesn't just answer questions; it gives sources and structured summaries. Each has its strengths, and I rotate between them depending on the task. For example, I might use Claude to summarise a report, Copilot to draft slides in PowerPoint or run data analysis in Excel, ChatGPT to rewrite it for different audiences, and Perplexity to validate the data and find citations. ## Choosing the right model for the job Each platform offers multiple models, fast, powerful, and sometimes ultra-powerful. For real work, you'll want to switch to the most capable version available: - **ChatGPT**: use GPT-4o (or GPT-o3 for advanced reasoning and deep outputs) - **Claude**: switch to Opus - **Microsoft Copilot**: built on OpenAI's models, typically running GPT-4, and optimised for tasks inside Word, Excel, PowerPoint, Outlook and Teams - **Perplexity**: uses strong retrieval-augmented models by default, so you're usually good to go Free versions are fine for testing, but if you're using AI seriously, expect to pay around **US$20–40/month** to unlock full capabilities. Privacy-wise, **Claude** is the most conservative, it never trains on your data. For **ChatGPT** and **Copilot**, you can turn off training in the settings (in Copilot's case, business users are usually covered by enterprise-grade data policies). I always disable training where I can. ## Here are the four characteristics of a business processes ripe for AI automation Over time, I've learned to identify business processes that AI can dramatically enhance. Here's the framework I share with clients: **Search** – When finding accurate information is slow or hard *- E.g. researching competitors, regulations, or market trends.* **Serve** – When your team needs to give instant, consistent answers *- E.g. providing product info to customers or contractors needing guidance during site inspections.* **Create** – When you're turning complex data into structured outputs *- E.g. writing RFPs, consent applications, or business reports.* **Enhance** – When business-critical decisions rely on fragmented data *- E.g. resource allocation, predictive maintenance, logistics planning.* ## The most underused superpower: Deep Research Most people don't realise AI can generate high-quality reports that feel like something a consultant would charge thousands for. I use Deep Research to: - Write strategic market overviews - Build competitive analysis decks - Generate CEO briefings on how AI is impacting their industry Perplexity is brilliant here because it cites its sources and lets you dive deeper. Claude and ChatGPT also do solid work, especially when you turn on web browsing. Gemini's reports can even turn into quizzes or podcasts with one click. ## Voice mode is more powerful than it looks I often use AI while walking or commuting. Voice mode in the ChatGPT and Gemini apps lets me have natural conversations and think through challenges aloud. But the real magic is **multimodal**, point your phone at a broken part, a whiteboard, a sign in another language, and the AI will help in real time. If you're just using voice mode like Siri, you're missing the best part. ## Making things with AI: code, docs, images, and video ChatGPT is my favourite for image generation and structured outputs like documents, dashboards, and workflows. Gemini is solid too, and Veo (its video model) creates surprisingly slick results with little effort. Claude can't make images, but it's excellent at writing, summarising, and analysing. Want AI to write a policy doc, turn it into a presentation, then create a social post and podcast episode about it? You can do that right now. ## Prompts aren't magic anymore, context is The key shift I see with today's models: it's less about the prompt and more about the **context**. Upload a doc, explain your situation, and be specific. Instead of "write me a job ad," say "write a job ad for a construction site supervisor in Auckland, here's our tone of voice, key responsibilities, and company story." Then refine from there. Ask for variations. Explore options. Use branches. Push it to do more, it doesn't get tired. ## My recommendation: spend your next hour like this Here's how I onboard new clients or team members into AI: - **Pick a system** and pay for it, the free versions are demos. - **Switch to the powerful model**, then give it a real task from your job. Add full context. Ask for changes until you love the output. - **Try Deep Research** on something you care about, a strategic issue, a product launch, or a market you're entering. - **Experiment with voice mode** while doing something else. Let the AI think with you in real time. Most people use AI like Google: quick, surface-level, disposable. But when you give it depth, real work, real context, it'll surprise you. ## Power up your potential with practical AI skills. Contact us to discuss how Artificial Intelligence could boost your business. [Book an AI Chat](https://www.newzealand.ai/contact-us/?service=ai-chat#book) [Explore the AI Operating System](https://www.newzealand.ai/ai-os/) --- # What comes after LLMs now has a New Zealand face on it. Source: https://www.newzealand.ai/what-comes-after-llms-now-has-a-new-zealand-face-on-it/ 22 June 2026 · Insights # What comes after LLMs now has a New Zealand face on it. **Justin Flitter** · Founder, AI New Zealand · 22 June 2026 · 3 min read *Two of the world's most respected AI researchers just raised around US$2 billion to build what comes after large language models. One of the best funded teams chasing the same idea is led by a Kiwi from Auckland.* Within weeks of each other, Yann LeCun and Fei-Fei Li raised about US$2 billion combined for the same contrarian bet. Not bigger LLMs. Not more parameters. Not longer context windows. Something fundamentally different. LeCun, who spent more than a decade as Meta's chief AI scientist, co-founded Paris-based AMI Labs and raised US$1.03 billion, the largest seed round in European history. Li, the Stanford professor behind ImageNet, raised US$1 billion for World Labs, backed by Nvidia, AMD and Andreessen Horowitz. Both are building world models: AI that learns from video, observation and physical reality and builds an internal picture of how the world actually works, rather than predicting the next word from text scraped off the internet. The thesis is simple. LLMs are brilliant with language but have no grounding in reality. They do not understand cause and effect, physics or consequences. World models are the attempt to fix that, and serious money now believes it is the next layer of the stack. Here is the part that should make every Kiwi sit up. The "what comes after LLMs" bet now has a New Zealand face on it. Odyssey, a London-based world model company, just closed a US$310 million (NZ$529m) Series B at a US$1.45 billion valuation. The round was led by Natural Capital, with Amazon, AMD Ventures, Google Ventures, In-Q-Tel and New Zealand's own Icehouse Ventures all in. Odyssey's co-founder and CTO is Jeff Hawke, a University of Auckland graduate who cut his teeth at self-driving startup Wayve, itself founded by another New Zealander, Alex Kendall. Two Kiwis at the frontier of teaching machines to learn from the real world. "A lot of AI companies are building tools on top of the big models," said Icehouse Ventures partner Mason Bleakely. "Odyssey is building one of those models for the physical world, which is a completely different level of ambition." In the same week, 21-year-old Thomas Mosen from Coromandel raised US$25 million from Caffeinated Capital for Sea12, his New York based back-office AI platform serving steel, construction and logistics firms. He started with US$30,000 and two interns in May 2024. Today Sea12 runs seven-figure revenue with 14 staff. Two Kiwis, two very different bets. Odyssey is wagering on a new architectural layer, world models trained on simulation, where the upside is robotics and physical infrastructure and so is the capital intensity. Sea12 is wagering on workflow specificity, that a steel fabricator will pay real money to automate the document chaos generic AI tools cannot reach. Both theses are credible. They are simply measured on completely different timescales. The lesson for New Zealand business is not that we should all rush off to build foundation models. It is that the talent shaping what comes next is ours. We are not just consumers of this technology. Increasingly, we are helping build it. **Sources** - Yann LeCun's AMI Labs raises US$1.03B to build world models, TechCrunch: [https://techcrunch.com/2026/03/09/yann-lecuns-ami-labs-raises-1-03-billion-to-build-world-models/](https://techcrunch.com/2026/03/09/yann-lecuns-ami-labs-raises-1-03-billion-to-build-world-models/) - **Fei-Fei Li's World Labs lands US$1B, TechCrunch:** [https://techcrunch.com/2026/02/18/world-labs-lands-200m-from-autodesk-to-bring-world-models-into-3d-workflows/](https://techcrunch.com/2026/02/18/world-labs-lands-200m-from-autodesk-to-bring-world-models-into-3d-workflows/) - World model maker Odyssey nabs US$1.45B valuation, TechCrunch: [https://techcrunch.com/2026/06/17/world-model-maker-odyssey-nabs-1-45b-valuation-backed-by-amazon-and-other-big-names/](https://techcrunch.com/2026/06/17/world-model-maker-odyssey-nabs-1-45b-valuation-backed-by-amazon-and-other-big-names/) - Kiwi founder Jeff Hawke secures $529m for Odyssey AI, NZ Herald: [https://www.nzherald.co.nz/business/kiwi-founder-jeff-hawke-secures-529m-for-odyssey-ai-with-backers-including-amazon-icehouse-ventures/2G42UFXPPNGYFKFM43WR6XWNTU/](https://www.nzherald.co.nz/business/kiwi-founder-jeff-hawke-secures-529m-for-odyssey-ai-with-backers-including-amazon-icehouse-ventures/2G42UFXPPNGYFKFM43WR6XWNTU/) - Coromandel's Thomas Mosen lands US$25m for Sea12, NZ Herald: [https://www.nzherald.co.nz/business/coromandels-thomas-mosen-lands-43m-backing-for-ai-start-up-sea12-in-us-boom/premium/AIIGQULDDNBCTHITBFOFEL7IY4/](https://www.nzherald.co.nz/business/coromandels-thomas-mosen-lands-43m-backing-for-ai-start-up-sea12-in-us-boom/premium/AIIGQULDDNBCTHITBFOFEL7IY4/) ## Power up your potential with practical AI skills. Contact us to discuss how Artificial Intelligence could boost your business. [Book an AI Chat](https://www.newzealand.ai/contact-us/?service=ai-chat#book) [Explore the AI Operating System](https://www.newzealand.ai/ai-os/) --- # What is 'Authenticity' in the AI era? Source: https://www.newzealand.ai/what-is-authenticity-in-the-ai-era/ 2 September 2025 · Gen AI # What is 'Authenticity' in the AI era? The traditional gatekeeping around "authentic" creativity is being challenged. **Justin Flitter** · Founder, AI New Zealand · 2 September 2025 · 3 min read Last week I commented on an RNZ radio interview with Jesse Mulligan; where a listener said that because someone read a poem at a wedding that was written by AI, it didn't have any soul. OUCH! My comment was It didn't matter if it was written by them, a random person or an AI tool. That the 'soul' of art or poetry is experienced by the consumer. If they find the poem, art or story connects, resonates with them then; it will have soul. The act of writing the content is not the act of giving something soul. **The soul lives in the connection, not the creation process** When someone at a wedding hears the AI-written poem and feels moved, laughs, or tears up, that emotional response is completely genuine. The poem did its job, regardless of its origin. It's a bit like asking whether a beautiful sunset is less meaningful because we understand the physics behind it. This reminds me of how we've always borrowed, adapted, and built upon existing ideas in creative work. Shakespeare lifted plots, musicians sample beats, and poets have always drawn from shared human experiences and language patterns. AI is just another tool in this continuum. **The real authenticity lies in the choice and curation** What I find exciting is that choosing the right AI-generated content, editing it, and deciding to share it at a meaningful moment, that's where human authenticity shines through. The person who selected that poem for the wedding made an authentic choice about what would resonate with their audience. We're moving towards a world where authenticity will be measured by emotional truth and connection rather than the mechanics of creation. And honestly, that feels like a more human way to think about art and expression. **How AI scales authenticity.** I want to share two practical ways your business can create content at scale that maintains your brand authenticity. **HeyGen: Scaling authenticity with AI** [HeyGen](https://email.email.circle.so/c/eJwczr2O6yAQQOGngc4Ihp_BBcVt_BrRMEyukbC9Sryxsk-_yhanPNJX6eT1Jhv1ceutZNcyeLQTxUhTQMSJLMjkHHmJFe7AQfOxbd97P9-fAxI4iLoV7xLFqKU4jAlTTJj1Whhqblx9axZxvtdP5BEkY5bMrHsBC9HO1gEARjSRXc4QmGaLlFpVwf7xDPcHDzHPQ4-ynufXU_l_ChYFy3VdZpX3f9kNH5uCpe-vfoqCJVkXqguiH2WVMQ4V7C7Xj9CgvRnq-lXgNwAA__8p_U0H) empowers organisations to create video content that feels personal and credible, at scale, across over 170 languages. Its AI Studio lets users start with a script and guides every aspect of their video, from emotional tone to gestures and voice, enabling brands to create presenters that truly mirror their values. With avatar-led video and seamless data integrations, HeyGen simplifies the workflow so brands can deliver handcrafted experiences to thousands without sacrificing warmth or sincerity. Brands like Skinny Mobile, OpenAI, Pepsico, and Samsung trust HeyGen to maintain their personality while reducing production time and costs. **Leveraging data to amplify personality** Scaling authenticity isn't just about automation, it's about strategic use of business data. Brands that respond personally to every customer inquiry build a foundation of communication that reflects their values and voice. By using this rich archive of personally written interactions, businesses can train AI systems to answer inquiries with the same warmth and style that defines their brand, scaling up without losing their unique resonance. Integrating customer data and workflows with AI allows companies to automate tasks while maintaining personalisation. The key is to treat the data not as a replacement for authenticity but as fuel, informing each interaction so it feels personal, connected, and valuable for every customer. **The future: human connection, powered by AI** When technology amplifies the best of a brand's people, authenticity becomes the outcome rather than the challenge. By blending personal communications with clever data integration, brands can create scale without sacrificing that vital trust and emotional connection that sets them apart. This shift isn't just about keeping up; it's leading the way for a more genuine, customer-centric future. ## Power up your potential with practical AI skills. Contact us to discuss how Artificial Intelligence could boost your business. [Book an AI Chat](https://www.newzealand.ai/contact-us/?service=ai-chat#book) [Explore the AI Operating System](https://www.newzealand.ai/ai-os/) --- # What is this 'MCP' thing AI experts are talking about? Source: https://www.newzealand.ai/what-is-this-mcp-thing-ai-experts-are-talking-about/ 30 March 2025 · Basics # What is this 'MCP' thing AI experts are talking about? MCP (Model Context Protocol) is a groundbreaking standard that acts as a universal connector between AI systems and the tools/data they need to interact with. **Justin Flitter** · Founder, AI New Zealand · 30 March 2025 · 2 min read Think of it as a "USB for AI" – a common interface that lets any AI model seamlessly access databases, apps, APIs, and other systems without custom coding. **What MCP Does** - **Universal Tool Access**: Allows AI to dynamically discover and use external resources like Google Drive, Slack, or coding environments through standardized connections - **Automatic Integration**: AI agents automatically detect available tools via MCP servers – no manual coding required (e.g., new CRM system appears as available tool immediately - **Vendor Agnostic**: Works with any AI model (Claude, GPT-4, open-source LLMs) and any service that implements MCP **Why It Matters for Business** - **Reduces Integration Costs**: Instead of building custom connections for each tool, companies use pre-built MCP servers (1,000+ available as of 2025) - **Enables Complex Workflows**: AI can chain actions across multiple systems – email a client update spreadsheet create Jira ticket, all through MCP - **Future-Proofs AI Investments**: Adopted by major tools like Replit, Codeium, and IDEs, MCP ensures AI capabilities grow as new tools join the ecosystem **Real-World Impact** Companies like Block (Square) use MCP to connect financial data systems, while design teams leverage it for AI-powered asset generation directly from Figma files[3][6]. This shifts AI from isolated chatbots to fully integrated business assistants that understand your specific tools and data context. **The protocol is particularly valuable for:** - **Non-technical users**: Claude Desktop app lets anyone use MCP-powered tools through natural language - **Cross-platform automation**: Mix AI providers (Claude + open-source models) while maintaining tool integrations - **Rapid prototyping**: Test workflows by asking AI to execute sequences in tools like Unity/Blender without manual setup MCP represents a fundamental shift in AI architecture – rather than just improving model intelligence, it focuses on connecting that intelligence to the real-world systems where work actually happens. For businesses, this means AI that truly understands your operations rather than existing in a vacuum. ## Power up your potential with practical AI skills. Contact us to discuss how Artificial Intelligence could boost your business. [Book an AI Chat](https://www.newzealand.ai/contact-us/?service=ai-chat#book) [Explore the AI Operating System](https://www.newzealand.ai/ai-os/) --- # When AI teaches itself: why digital minds learn differently Source: https://www.newzealand.ai/when-ai-teaches-itself-why-digital-minds-learn/ 8 March 2026 · Basics # Stepbystep logic **Justin Flitter** · Founder, AI New Zealand · 8 March 2026 · 3 min read Humans and AI are both neural systems, but they solve *different* learning problems because they have opposite constraints: humans have huge capacity and little data; big neural nets have smaller capacity and massive data. *These are some of my notes from listening to this interview with* [*Geoffrey Hinton*](https://www.youtube.com/watch?v=l6ZcFa8pybE) *on Sunday. It's well worth a watch to hear the godfather of AI explain how the AI models of today work.* - The human brain has about 100 trillion connections (synapses), but a person only lives for a few billion seconds, so we get relatively **few** experiences compared with our capacity. - Large AI models have far fewer connections than us (around 1% of our synapses, roughly a trillion parameters), but they are trained on *thousands of times* more data than a single human ever sees. - Backpropagation is very good at compressing huge amounts of experience into a limited number of connections, which suits AI's "lots of data, not as many connections" regime. - **Humans face the opposite regime:** "lots of connections, not much data", so our brains must extract as much as possible from each experience, likely using learning mechanisms that are not plain backpropagation. ## Scaling and selfgenerated data - When engineers made neural nets bigger and fed them more data, performance improved in a predictable way, as long as they didn't run out of data. - For many tasks, you eventually hit a wall because there is only so much highquality human data available; beyond that, making models bigger gives diminishing returns. - Some systems dodge this limit by generating their **own** data through selfplay or simulation, like AlphaGo/AlphaZero playing games against themselves to keep improving far beyond human experts. ## Applying this to language models - **Today's language models mainly learn by imitating humans:** they predict the next word in humanwritten text, analogous to mimicking expert moves in Go; this caps performance at roughly "expert level". - A more powerful approach is for models to use reasoning over their own beliefs expressed in language: derive new conclusions, detect inconsistencies, and then adjust either their beliefs or their reasoning process. - If a model can iteratively notice "my beliefs contradict each other, so something must change", it effectively generates its own learning signals, just like selfplay generates its own game data. ## Intelligence, creativity, and experience - With selfgenerated "mental data" of this kind, language models could become much more intelligent than humans, at least in reasoning and knowledge organisation. - However, creating works that deeply matter to humans (for example, a Shakespearelevel novel) might still require experiences similar to ours, such as embodiment and genuine exposure to human life. ## Mortality and digital minds - Human minds are "analog" in the sense that when the brain dies, its exact synaptic configuration is gone, so our knowledge dies with us. - **Digital intelligences are different:** you can save their parameters, destroy all hardware, then later rebuild hardware and reload those weights so the same mind "comes back to life". - In this sense, digital systems have a practical form of resurrection: their identity and knowledge can be paused and restarted indefinitely, unlike human beings. ## Now, the question is: *Can language models do the same thing?* Currently, models learn by predicting the next word in human text, just as AlphaGo once mimicked experts. To surpass humans, language models would need to **generate and test their own beliefs**, not just copy ours. A model could take what it already believes (its internal "knowledge"), reason about it, and detect inconsistencies - e.g., "If A and B are true, then C should be true... but I don't believe C." Resolving such conflicts would let the model correct itself and grow smarter - without more external data. This would be the **next stage of AI learning**: - Internally generate reasoning questions (selfplay for thought). - Detect contradictions. - Revise internal models through reasoning. That approach blends deep learning's pattern recognition with symbolic reasoning's logic - creating a selfimproving system. Models like **Gemini** are already believed to be moving in this direction. ## Power up your potential with practical AI skills. Contact us to discuss how Artificial Intelligence could boost your business. [Book an AI Chat](https://www.newzealand.ai/contact-us/?service=ai-chat#book) [Explore the AI Operating System](https://www.newzealand.ai/ai-os/) --- # Why AI Models Hallucinate and How to Prevent It. Source: https://www.newzealand.ai/why-ai-models-hallucinate-and-how-to-prevent-it/ 16 October 2025 · Guidelines # Why AI Models Hallucinate and How to Prevent It. AI hallucinations are one of the most fascinating challenges we're tackling in this space. The good news? **Justin Flitter** · Founder, AI New Zealand · 16 October 2025 · 6 min read Understanding why they happen is the first step to working around them effectively. ## Why AI Models Hallucinate AI hallucinations occur when models generate plausible-sounding but incorrect, misleading, or fabricated information. Unlike human hallucinations caused by brain disorders, AI hallucinations stem from the fundamental way these systems work, they're pattern-prediction machines, not knowledge databases. **The Core Problem: Guessing Over Uncertainty** OpenAI's latest research reveals that language models hallucinate because their training and evaluation procedures reward guessing over acknowledging uncertainty. When an AI doesn't know something, it doesn't say "I don't know", instead, it predicts the most statistically probable next word based on patterns it learned, even when that leads to fabrication. **Root Causes Include:** **Insufficient or flawed training data**: If the training dataset is incomplete, biased, outdated, or contains inaccuracies, the model learns incorrect patterns. The quality of outputs is directly tied to the quality of inputs used during training. **Lack of real-world grounding**: Most models generate outputs based on training data patterns rather than accessing verified knowledge bases, meaning they can't distinguish between what's plausible and what's true. **Probabilistic nature**: AI predicts sequences based on statistical likelihood, not factual accuracy. It generates what sounds right, not necessarily what is right. **Overfitting**: When models memorise specific training examples rather than learning generalisable patterns, they apply irrelevant or inappropriate information to new contexts. **Model complexity**: Highly complex models without proper constraints can produce a wider range of outputs, increasing the likelihood of hallucinations. **Encoding and decoding errors**: Mistakes in how the model processes information, learning wrong correlations or attending to the wrong parts of input can produce hallucinated outputs. ## Scenarios and Behaviours That Trigger Hallucinations Understanding when hallucinations are most likely helps you avoid them proactively. **High-Risk Scenarios:** **Vague or ambiguous prompts**: Unclear instructions force the model to make assumptions, leading to generic or incorrect responses. Asking "Tell me about space" versus "Summarise NASA's recent Mars missions with factual details from official reports" produces vastly different reliability levels. **Complex or multi-step queries**: Cramming multiple topics into one prompt increases confusion and inaccurate outputs. Breaking requests into focused, sequential prompts dramatically improves accuracy. **Domain-specific or specialised knowledge**: Hallucination rates vary by domain. Medical information averages 4.3-15.6% hallucination rates, legal information 6.4-18.7%, and scientific research 3.7-16.9%. **Requests for citations or sources**: When asked for references, AI may fabricate plausible-sounding but completely nonexistent citations, articles, or legal cases. One study found ChatGPT falsely attributed 76% of quotes from journalism sites. **Adversarial or nonsensical prompts**: Deliberately confusing inputs or prompts composed of random tokens can trigger hallucinations. This reveals that hallucinations share characteristics with adversarial examples in machine learning. **Idioms and slang**: Expressions the model wasn't trained on can lead to nonsensical outputs. **Out-of-date information**: When models lack current data, they rely on outdated training information, producing factually incorrect responses. **Using external tools**: LLM agents augmented with external tools can exhibit significantly higher hallucination rates due to added complexity. **Temperature Settings:** The temperature parameter controls randomness in outputs. Higher temperatures (0.7-1.0) increase creativity but also hallucination risk, while lower temperatures (0-0.3) produce more focused, factual responses. However, setting temperature to zero doesn't eliminate hallucinations, it only makes outputs deterministic. **Cascade Effect:** In conversational AI, each generated word influences the next. If the model starts with a hallucination, it can build upon that error, creating increasingly inaccurate responses as the conversation continues. ## Best Practice Instructions to Reduce Hallucinations Here's your practical playbook for minimising AI hallucinations in business applications: ## 1 Use Retrieval-Augmented Generation (RAG) RAG is the gold standard for reducing hallucinations. It connects AI models to external, verified knowledge bases, like your company's documentation, trusted databases, or authoritative sources, before generating responses. Instead of relying solely on training data, the model retrieves relevant information from trusted sources first, then generates answers grounded in that data. Benefits: Reduces hallucinations, eliminates expensive retraining, provides source citations users can verify, and keeps information current. ## 2 Engineer Clear, Specific Prompts Precision in prompting is non-negotiable: - **Be specific**: Replace "Tell me about marketing" with "What are three specific strategies to increase customer retention for B2B SaaS companies in 2025?" - **Provide context**: Include relevant background information, audience details, and desired format - **Break down complex requests**: Split multi-part questions into sequential, focused prompts - **Set clear expectations**: Use specific language that guides the model toward known data sources - **Request verification**: Ask the AI to cite sources or explain its reasoning step-by-step - **Specify constraints**: Define word count, format, time frame, or other boundaries to focus outputs ## 3 Use Chain-of-Thought (CoT) Prompting Ask models to break down their reasoning step-by-step before arriving at answers. This technique improves accuracy by 35% in reasoning tasks and reduces mathematical errors by 28%. Example: "Break down the steps to calculate 17 multiplied by 24 before giving the final answer." ## 4 Implement Chain-of-Verification (CoVe) This four-step process reduces hallucinations: - Generate a baseline response - Create verification questions to test accuracy - Execute verification by checking responses against original answers - Generate final response with corrections applied ## 5 Adjust Temperature Settings Appropriately Use low temperature settings (0-0.3) for factual tasks requiring accuracy. Use higher temperatures (0.7-1.0) only for creative, open-ended tasks like brainstorming where factual precision is less critical. ## 6 Establish Human-in-the-Loop (HITL) Processes 76% of enterprises now include human review processes to catch hallucinations before deployment. Human oversight provides essential fact-checking that AI cannot perform on its own. Knowledge workers currently spend an average of 4.3 hours per week fact-checking AI outputs. ## 7 Fine-Tune on High-Quality Data When you have standardized tasks and sufficient training data, fine-tuning models on curated, accurate, domain-specific information significantly reduces hallucinations. Ensure data comes from verified, reputable sources with biases removed. ## 8 Implement Post-Processing Verification Compare AI-generated outputs against reliable factual databases. Cross-reference information with trusted sources like academic databases, official government sites, or peer-reviewed publications. Use fact-checking tools and always verify citations actually exist. ## 9 Direct AI to Acknowledge Uncertainty Instruct models that "no answer is better than an incorrect answer". This reduces the likelihood of fabricated responses when the AI lacks sufficient information. ## 10 Use Constrained Prompting Limit the model's output by providing strict guidelines on format, length, tone, and acceptable sources. Example: "Using only information from Wikipedia, provide three bullet points explaining..." ## 11 Provide Examples in Prompts (Few-Shot Learning) Including 2-3 correct examples in your prompt guides the AI toward the type of information you're requesting. ## 12 Use Trusted, Domain-Specific Models Generic models like ChatGPT work for general tasks, but domain-specific models trained on specialised datasets produce fewer hallucinations in their areas of expertise. Consider using models specifically designed for your industry or use case. ## 13 Maintain Up-to-Date Knowledge Bases If using RAG, ensure your external data sources are regularly updated through automated real-time processes or periodic batch processing. ## 14 Avoid Information Overload Don't cram excessive information into single prompts. This causes confusion and incoherent responses. Focus each prompt on a single, manageable task. ## 15 Establish Verification SOPs Create standard operating procedures for consistent review, including what to check, how to correct errors, and how to flag edge cases. Train teams on interpreting AI results and when to intervene. ## The Reality Check Even with best practices, hallucinations cannot be completely eliminated with current AI technologies. Leading models like Google Gemini-2.0-Flash-001 achieve hallucination rates as low as 0.7-0.9%, while most widely-used models fall between 2-5%. Paradoxically, more advanced reasoning models sometimes hallucinate more frequently, OpenAI's o3 and o4-mini hallucinated 33% and 48% respectively on certain benchmarks. The key is implementing layered safeguards: RAG for grounding, precise prompting for clarity, verification processes for accuracy, and human oversight for accountability. This combination creates trustworthy AI systems suitable for business applications. The opportunity here is massive. Organisations that master hallucination mitigation will build more reliable AI systems, earn greater user trust, and unlock applications in sensitive domains like healthcare, finance, and legal services. Understanding these challenges positions your consultancy to guide New Zealand businesses toward AI implementations that actually deliver on their promises. ## Power up your potential with practical AI skills. Contact us to discuss how Artificial Intelligence could boost your business. [Book an AI Chat](https://www.newzealand.ai/contact-us/?service=ai-chat#book) [Explore the AI Operating System](https://www.newzealand.ai/ai-os/) --- # You Don't Have to Rent Your AI Source: https://www.newzealand.ai/you-don-t-have-to-rent-your-ai/ 17 June 2026 · Guidelines # First, untangle what "open source" means **A practical guide to the open-source tools that do what ChatGPT and Claude do, using open models you can run on your own terms.** **Justin Flitter** · Founder, AI New Zealand · 17 June 2026 · 5 min read Most Kiwi businesses meet AI through a rented front door. You log into ChatGPT or Copilot, you type, you get an answer, and your data takes a quiet trip offshore to someone else's servers. That is fine for a lot of work. It is not the only option, and for a growing number of teams it is no longer the smartest one. There is a whole category of tools that look and feel like ChatGPT, let you create content and search the web, but run on open models you can host yourself. Some are free. Some are genuinely open source. A few are neither, and we will be honest about which is which. Here is the practical landscape, sorted by what you actually want to do. This is where most articles get lazy, so let us be precise. Three different things hide under the same phrase. **Open-weight models** are the brains. Meta's Llama, Mistral, DeepSeek, Alibaba's Qwen, Google's Gemma and OpenAI's GPT-OSS are released for anyone to download and run. In 2026 the best of them rival the big proprietary systems on everyday business tasks like drafting, summarising and research. **Open-source apps** are the interface, the bit that looks like ChatGPT. Tools like Open WebUI, Jan and AnythingLLM are free to inspect, modify and self-host. **Free but proprietary** tools sit in the middle. They cost nothing and run open models, but you cannot see under the bonnet. LM Studio is the obvious example. Keep those three layers separate and the rest of this makes sense. ## If you want to try open models with zero setup Start in your browser. No install, no hardware, no commitment. [**HuggingChat**](https://huggingface.co/chat) is the easiest on-ramp. Hugging Face briefly retired the original last year and relaunched it as HuggingChat Omni, which routes your prompt to a suitable open model automatically instead of making you choose. It is free, and it is the fastest way to feel how good open models have become. [**Le Chat**](https://chat.mistral.ai), from French company Mistral, is the most polished hosted option. It does web search, image generation, document handling and has a no-code agent builder. Mistral is an open-weight model maker, and Le Chat is their assistant sitting on top. For teams nervous about US data handling, a European provider is a meaningful difference. [**OpenRouter**](https://openrouter.ai) gives you one login to dozens of models, including free open ones, so you can compare answers side by side before committing to anything. A caution worth stating plainly. [DeepSeek](https://chat.deepseek.com) and [Qwen](https://chat.qwen.ai) both offer cheap, capable chat apps, and both are hosted in China. The models are excellent and you can run them privately yourself. Consider here, where your client data or anything commercially sensitive should go. Under the Privacy Act 2020, where personal information ends up is your responsibility, not the vendor's. ## If you want AI running on your own laptop This is the privacy play. Nothing leaves your machine, there are no per-seat fees, and it works on a plane. [**Jan**](https://jan.ai) is the one to start with. It is properly open source, runs fully offline once you have downloaded a model, and is built privacy-first. It will also connect out to ChatGPT or Claude if you want, so you are not locked into local-only. [**LM Studio**](https://lmstudio.ai) is the slickest experience and the gentlest learning curve. Worth repeating the nuance from earlier: it is free, including for work, but it is proprietary software, not open source. If that distinction matters to you, it should inform the choice. If it does not, LM Studio is a lovely two-click way in. Underneath a lot of these sits [**Ollama**](https://ollama.com), the engine that actually runs the model. You can pair it with a tidy interface, or let tools like Jan and Open WebUI handle that for you. [**GPT4All**](https://www.nomic.ai/gpt4all) and [**Msty**](https://msty.app) are two more friendly desktop apps worth a look. ## If you want to roll it out to a team This is where open source starts to pay rent. Self-host once, and the whole team gets a private ChatGPT with no monthly per-user bill. [**Open WebUI**](https://openwebui.com) is the de facto standard. It is one of the most popular self-hosted AI interfaces going, with user accounts, chat history, web search, and the ability to chat with your own documents. It plugs into Ollama or any model you like. You will need someone comfortable with Docker to stand it up, but once it is running the experience rivals the commercial products. [**AnythingLLM**](https://anythingllm.com) is the document-first option, open source, and runs as a desktop app or on a server. [**LibreChat**](https://www.librechat.ai) unifies multiple providers in one interface and adds web search and tool use. All three keep your conversations and files on infrastructure you control. ## If you want AI-powered web search you can host Perplexity-style "ask a question, get a cited answer" search has an open-source answer too. [**Perplexica**](https://github.com/ItzCrazyKns/Perplexica), recently rebranded **Vane**, runs entirely on your own hardware. It pairs a private search engine with the model of your choice and returns answers with sources, without routing your queries through a vendor's black box. [**Morphic**](https://morphic.sh) is a second open answer engine you deploy yourself. Both give you cited, current answers while keeping the questions private. ## The honest trade-off Open and local is not free in every sense. You trade some convenience for control. Hosted tools are effortless but your data leaves the building, so check where it lands. Self-hosted tools keep everything in-house but need a bit of setup and decent hardware. There is no free lunch, only a choice about what you are optimising for: ease, cost, or sovereignty. The teams that get value are not the ones with the cleverest tool. They are the ones who picked a real problem first. So move a metric, not a use case. ## Where to start next week You do not need a project plan. You need an experiment. - **Pick one low-stakes workflow** where output is slow or inconsistent. Drafting proposals, summarising long documents, first-pass research. Name the lever you want to move: cycle time or cost-to-serve. - **Spend 30 minutes in HuggingChat or Le Chat.** Feel the quality of today's open models on your real work, not a demo prompt. - **If privacy is the blocker, install Jan or LM Studio** and run a model fully offline. Notice that nothing leaves your laptop. That changes the conversation about what AI is allowed to touch. - **Ring-fence a small monthly slot** for one person to test a tool and report back. Curiosity is the antidote to obsolescence, and it compounds when it is shared. - **Keep humans in the loop.** AI drafts, humans approve. Open models are powerful, not infallible. Open source has quietly closed the gap. The models are good, the tools are mature, and for the first time you can have the ChatGPT experience without renting it, without the lock-in, and without your data leaving the country. That is not a downgrade. That is control. ## Power up your potential with practical AI skills. Contact us to discuss how Artificial Intelligence could boost your business. [Book an AI Chat](https://www.newzealand.ai/contact-us/?service=ai-chat#book) [Explore the AI Operating System](https://www.newzealand.ai/ai-os/) --- # Your People Are Ready for AI. Your Infrastructure Isn't. Source: https://www.newzealand.ai/your-people-are-ready-for-ai-your-infrastructure-isn-t/ 19 March 2026 · Guidelines # AI Doesn't Fix Broken Plumbing. It Floods the House. The invisible foundation every business needs before AI can deliver real value. **Justin Flitter** · Founder, AI New Zealand · 19 March 2026 · 5 min read Most businesses talking about AI are not short on ambition. They're short on infrastructure. Not servers and GPUs, the operational infrastructure that determines whether information flows or gets stuck. Document management that lives in people's heads. Data trapped in inboxes. Meeting decisions that never make it into any system. Standard operating procedures that don't exist, or exist but nobody follows. **This is the real AI readiness gap. And no amount of prompt engineering will close it.** ## What's Changing, and What's at Risk - **Data silos are not just inconvenient, they're expensive.** IDC research shows companies lose 20–30% of revenue annually to inefficiencies caused by disconnected systems. McKinsey estimates data silos cost businesses US$3.1 trillion globally in lost productivity. - **Shadow AI is already inside your organisation.** 68% of employees use free-tier AI tools via personal accounts, with 57% inputting sensitive company data. That's not innovation, it's uncontrolled IP leakage. - **Manual information processing is a hidden tax.** Workers lose an average of 12 hours per week searching for information trapped in silos. People hired simply to export data from one system, modify it, and import it into another, a never-ending cycle of copy, paste, reconcile. - **Meeting intelligence is leaking out the door.** Notes jotted on paper. Free transcription tools sending recordings to third-party servers that may train public AI models. Vendor terms that allow data to be shared with affiliates. Every unmanaged meeting tool is a confidentiality risk. - **AI amplifies whatever it finds.** If your data is scattered, inconsistent, and ungoverned, AI will generate scattered, inconsistent, and ungoverned outputs, faster. AI doesn't fix broken processes. It accelerates them. **The question is not "Which AI tool should we buy?" It's "Is our organisation structured so that AI can actually find, trust, and use our information?"** ## The Infrastructure Gap, What Businesses Get Wrong When leaders say "infrastructure" they think about IT hardware. That's not the infrastructure holding most businesses back. The real gaps are operational: ## 1. No single source of truth Critical documents, contracts, invoices, proposals, standard operating procedures, live in Outlook inboxes, personal drives, and desktop folders. Nobody can find the latest version. Nobody knows which version is authoritative. When AI agents try to retrieve information, they find nothing, because the information isn't in any system AI can access. ## 2. No document and data management standards Employees aren't saving files to SharePoint or a centralised system because no one has told them to. There's no standard naming convention, no folder structure, no retention policy. It's not a technology problem, it's a governance gap. And it means your organisational knowledge is held hostage by individual habits. ## 3. Cobbled-together systems with disparate data The typical mid-size business runs dozens of disconnected apps. CRM, accounting, project management, HR, email, file storage, none of them talking to each other. People become the integration layer. Their job is to be the human API: export from system A, reformat in Excel, import into system B. It's fragile, error-prone, and completely invisible on any balance sheet. ## 4. Meeting intelligence disappears Decisions made in meetings rarely make it into a system of record. Handwritten notes get lost. Free transcription tools create transcripts that sit on third-party servers, disconnected from your organisational knowledge, exposed to vendor data-use policies, and unavailable when an AI assistant could actually use them. This is both an IP leakage risk and a knowledge management failure. ## 5. No governance for AI tool use Without sanctioned tools and clear policies, employees will, and do, use whatever free AI tool they can find. They paste customer data into ChatGPT. They upload contracts to unvetted summarisation tools. They don't do this maliciously. They do it because no one gave them a better option. ## The Four Foundations of AI-Ready Infrastructure Before any AI tool can deliver real business value, four operational foundations must be in place. Think of these as the plumbing, wiring, and building code that make the house functional, long before you choose the furniture. ## Foundation 1: Sense, One Source of Truth - Centralise documents in a governed system (e.g. SharePoint, Google Drive, or equivalent) with clear folder structures, naming conventions, and access controls. - **Automate extraction:** invoices, contracts, and key documents should flow into the system without relying on individuals to remember to save them. - Establish a retention and version control policy. If people can't trust the system has the latest version, they won't use it. ## Foundation 2: Think, Connected and Clean Data - Audit your systems. Map where data lives, how it moves between systems, and where humans are acting as the integration layer. - Prioritise integration over replacement. Connect your existing tools through automation platforms (e.g. Zapier, Power Automate, [make.com](http://make.com/)) before buying new ones. - Establish data quality standards. AI outputs are only as reliable as the data they draw from. Garbage in, garbage out, just faster. ## Foundation 3: Decide, Governance and Guardrails - Publish an AI acceptable use policy. Specify which tools are sanctioned, what data can and cannot be shared with AI, and who is responsible for reviewing AI-generated outputs. - Replace shadow AI with sanctioned alternatives. If your people need meeting transcription, give them a tool that keeps data inside your organisation, not one that sends it to a third party to train public models. - **Define decision rights:** what AI can draft, what humans must review, and what must never be delegated. ## Foundation 4: Act, Standard Operating Procedures - Document the workflows that matter most. Start with 3–5 high-volume, high-value processes and write them down. - Train your people. Not on AI tools, on the operational disciplines that make AI tools effective: where to save files, how to name them, when to use which system. - Build feedback loops. Measure whether the foundations are being followed, and refine them quarterly. ## Three Decisions You Need to Make This Quarter - **Appoint an owner for information architecture.** Not an IT project, a business accountability. Someone must own how your organisation stores, structures, and governs its documents and data. Without an owner, nothing changes. - **Audit and standardise your document management.** Pick your platform (most New Zealand businesses already have SharePoint or Google Drive). Define the folder structure, naming conventions, and access policies. Communicate them. Enforce them. This is the single highest-ROI action most organisations can take before investing in AI. - **Publish an AI governance policy.** Specify sanctioned tools. Ban the use of free-tier personal AI accounts for company data. Provide clear guidance on what information can be shared with AI and what cannot. Make the policy short, practical, and enforceable. ## AI commoditises answers. Human judgement becomes premium. But judgement needs information it can trust. And right now, most organisations can't even find theirs. Fix the infrastructure. Then let AI do what it does best. ## Power up your potential with practical AI skills. Contact us to discuss how Artificial Intelligence could boost your business. [Book an AI Chat](https://www.newzealand.ai/contact-us/?service=ai-chat#book) [Explore the AI Operating System](https://www.newzealand.ai/ai-os/) --- # AI Glossary: 100+ terms every NZ business leader needs Source: https://www.newzealand.ai/ai-glossary/ 3 February 2026 · Guidelines # AI Glossary: 100+ Terms Every NZ Business Leader Should Know. Most AI glossaries read like vendor brochures. Lots of definitions. Not much help when you're trying to figure out what's real, what's hype, and what actually matters for your business. **Justin Flitter** · Founder, AI New Zealand · 3 February 2026 · 31 min read This one is different. Every term here includes the definition *and* the NZ business context you need to interrogate claims, spot AI theatre, and ask better questions. Because in 2026, the competitive advantage isn't knowing what AI can do - it's knowing what questions to ask when someone tells you what AI *should* do. Critical thinking is your moat. This glossary is your toolkit. ## Foundation Concepts ### **AI (Artificial Intelligence)** **Definition** Systems that perform tasks typically requiring human intelligence, pattern recognition, decision-making, language processing, prediction. **NZ business context** When someone pitches "AI", ask: "Which specific task does this handle, and what judgement does my team still own?" AI is a category, not a capability. The question isn't whether to use AI, it's where human judgement stays in the loop and where automation actually moves a business metric. ### **Machine Learning (ML)** **Definition** A subset of AI where systems learn patterns from data without being explicitly programmed for every scenario. **NZ business context** ML needs clean, relevant data to learn from. For NZ businesses, this means: Do you have enough historical data? Is it representative of current conditions? Can you measure whether the model is learning the right patterns or just amplifying existing bias? ML is only as good as the data and the judgement applied to its outputs. ### **Generative AI** **Definition** AI systems that create new content, text, images, code, audio, based on patterns learned from training data. **NZ business context** Generative AI (like ChatGPT, Claude, Gemini) drafts, structures, and suggests. It doesn't verify, approve, or guarantee accuracy. For NZ businesses, this means establishing clear workflows: AI generates, humans review, humans decide. Never let generative AI operate without a review step in regulated, customer-facing, or high-stakes contexts. ### **Large Language Model (LLM)** **Definition** AI models trained on vast amounts of text data to understand and generate human-like language. Examples: GPT-4, Claude, Gemini, Llama. **NZ business context** LLMs are pattern-matching engines, not knowledge databases. They predict likely word sequences based on training data. Ask: "Is this answer contextually right for *our* business, or just statistically plausible?" LLMs can be confidently wrong. Your team's job is to catch it. ### **Natural Language Processing (NLP)** **Definition** The branch of AI focused on understanding, interpreting, and generating human language.\ **NZ business context** NLP powers chatbots, email sorting, sentiment analysis, and document summarisation. For NZ businesses, the key question is: "Does this understand our context, industry jargon, Kiwi English, regional nuances, or is it optimised for US/UK datasets?" Test before you trust. ### **Neural Network** **Definition** A computing architecture modelled loosely on the human brain, using interconnected nodes (neurons) to process information and learn patterns. **NZ business context** You don't need to build neural networks, but you need to know they underpin most modern AI. The practical question: "Can we explain how this system reached its conclusion?" Many neural networks are black boxes. If you're in a regulated industry (finance, health, insurance), explainability matters. ### **Deep Learning** **Definition** A subset of machine learning using multi-layered neural networks to process complex data like images, speech, and unstructured text. **NZ business context** Deep learning powers image recognition, voice assistants, and advanced diagnostics. It requires significant compute power and large datasets. For most NZ SMEs, you'll consume deep learning via APIs (Google Vision, AWS Rekognition) rather than building it in-house. Ask: "What's the cost, and does it move a lever?" ### **Algorithm** **Definition** A set of rules or instructions a computer follows to solve a problem or complete a task. **NZ business context** Every AI system runs on algorithms. The question isn't whether an algorithm is involved, it's whether the algorithm reflects your business logic, your risk tolerance, and your values. Algorithms can encode bias, prioritise the wrong outcomes, or optimise for metrics that don't matter. Interrogate the rules, not just the results. ### **Model** **Definition** The trained AI system that makes predictions, classifications, or generates outputs based on learned patterns. **NZ business context** A "model" is the output of training an AI system on data. For NZ businesses: Who trained it? On what data? When? Models trained on international datasets may not reflect NZ market conditions, customer behaviour, or regulatory requirements. Always test models in your context before deploying. ### **Training Data** **Definition** The dataset used to teach an AI model patterns, relationships, and behaviours. **NZ business context** Training data determines what the model learns. If your training data is biased, incomplete, or outdated, your model will be too. For NZ businesses, ask: "Does this data reflect our customers, our market, our operating environment?" And: "Who owns this data, and what are the privacy implications? ### **Inference** **Definition** The process of an AI model making predictions or generating outputs based on new input data after training is complete. **NZ business context** Inference is when the AI does its job, analysing a customer query, flagging a risk, drafting a response. For NZ businesses, the cost and speed of inference matter. Some models are expensive to run at scale. Ask: "What's the per-query cost, and does the value justify it?" ## Capabilities & Techniques ### **Prompt Engineering** **Definition** The practice of designing and refining input instructions (prompts) to get better outputs from AI models. **NZ business context** Prompt engineering is now a core professional skill. The quality of your question determines the quality of the answer. For NZ businesses, this means: Train your team to write clear, context-rich prompts. Test variations. Document what works. Poor prompts = poor outputs, even with powerful models. ### **Fine-Tuning** **Definition** The process of taking a pre-trained AI model and further training it on a specific dataset to improve performance for a particular task or domain. **NZ business context** Fine-tuning makes a general model more useful for your business - teaching it your industry terminology, your customer language, your processes. For NZ businesses, this requires clean domain-specific data and technical expertise. Ask: "Is fine-tuning necessary, or can we achieve the same result with better prompts and retrieval?" ### **Retrieval-Augmented Generation (RAG)** **Definition** A technique that combines AI generation with real-time retrieval of relevant information from a knowledge base or database, improving accuracy and reducing hallucinations. **NZ business context** RAG lets you ground AI outputs in your own documents, policies, and data - without retraining the entire model. For NZ businesses, this is often the most practical path to accurate, context-aware AI. Ask: "What's our source of truth, and how do we keep it current?" No citation, no trust. ### **Embeddings** **Definition** Numerical representations of words, sentences, or documents that capture semantic meaning, allowing AI to understand similarity and context. **NZ business context** Embeddings power search, recommendation, and retrieval systems. For NZ businesses, this means you can find documents, customer queries, or products based on *meaning*, not just keywords. The practical question: "Does this improve findability and relevance for our team and customers?" ### **Vector Database** **Definition** A database optimised for storing and querying embeddings (vectors), enabling fast semantic search and retrieval. **NZ business context** Vector databases underpin RAG systems and AI-powered search. For NZ businesses deploying internal knowledge systems or customer support AI, vector databases are the infrastructure. Ask: "What's the hosting cost, and who manages updates to the knowledge base?" ### **Tokenisation** **Definition** The process of breaking text into smaller units (tokens) that AI models can process. Tokens can be words, subwords, or characters. **NZ business context** Tokens determine cost and capacity. Most LLM pricing is per token (input + output). For NZ businesses, this means: Long documents = higher costs. Verbose prompts = higher costs. Ask: "What's our average token usage per query, and what's the monthly cost at scale?" ### **Context Window** **Definition** The maximum amount of text (measured in tokens) an AI model can process in a single interaction - input plus output. **NZ business context** Context windows limit how much information you can feed into an AI at once. As of early 2026, leading models support 128K–200K tokens (roughly 100–150 pages). For NZ businesses, this means you can process entire reports, contracts, or transcripts in one go - but you'll pay for it. Ask: "Do we need the full document, or can we retrieve relevant sections?" ### **Zero-Shot Learning** **Definition** An AI model's ability to perform a task it wasn't explicitly trained for, based on general knowledge and instructions. **NZ business context** Zero-shot means you can ask an LLM to do something new without providing examples. For NZ businesses, this is powerful for ad-hoc tasks - summarising a report, drafting an email, translating jargon. But: zero-shot outputs are less reliable than few-shot. Always review. ### **Few-Shot Learning** **Definition** Providing an AI model with a few examples of a task to improve its performance on similar tasks. **NZ business context** Few-shot prompts improve accuracy and consistency. For NZ businesses, this means: If you want AI to draft customer responses in your tone, show it 3–5 examples. If you want it to extract data in a specific format, provide sample outputs. Few-shot is your quality control lever. ### **Transfer Learning** **Definition** Reusing a model trained on one task to improve performance on a related task, reducing the need for large datasets and training time. **NZ business context** Transfer learning is why you don't need to train an LLM from scratch. Pre-trained models (GPT, Claude, Gemini) already understand language, reasoning, and general knowledge. For NZ businesses, you build on top of these models via prompts, fine-tuning, or RAG. Ask: "What's the fastest path to value, prompts, retrieval, or fine-tuning?" ### **Reinforcement Learning from Human Feedback (RLHF)** **Definition** A training method where human evaluators rank AI outputs, and the model learns to prioritise responses that align with human preferences. **NZ business context** RLHF is how models like ChatGPT learn to be helpful, harmless, and honest. For NZ businesses deploying custom AI, RLHF means: You can teach the model your preferences - tone, risk tolerance, compliance standards - by rating outputs. This requires ongoing human input and clear evaluation criteria. ### **Supervised Learning** **Definition** A machine learning approach where the model is trained on labelled data - input-output pairs where the correct answer is known. **NZ business context** Supervised learning powers predictive models - credit scoring, fraud detection, demand forecasting. For NZ businesses, this requires historical data with known outcomes. Ask: "Do we have enough labelled examples? Is the data representative of current conditions? Who labels new data as the business evolves?" ### **Unsupervised Learning** **Definition** A machine learning approach where the model identifies patterns, clusters, or anomalies in data without labelled examples. **NZ business context** Unsupervised learning is useful for exploratory analysis - customer segmentation, anomaly detection, trend identification. For NZ businesses, this can surface insights you didn't know to look for. But: the model finds patterns, not explanations. Your team still needs to interpret and validate findings. ### **Classification** **Definition** A machine learning task where the model assigns inputs to predefined categories (e.g., spam/not spam, high risk/low risk). **NZ business context** Classification powers email filtering, loan approvals, quality control, and triage systems. For NZ businesses, the key question: "What happens when the model gets it wrong?" Build review workflows for edge cases and high-stakes decisions. ### **Regression** **Definition** A machine learning task where the model predicts a continuous numerical value (e.g., sales forecast, customer lifetime value, delivery time). **NZ business context** Regression helps NZ businesses forecast demand, estimate costs, and predict outcomes. The trap: models assume the future resembles the past. If market conditions shift, your regression model may be confidently wrong. Always monitor actual vs. predicted and retrain when drift occurs. ### **Clustering** **Definition** An unsupervised learning technique that groups similar data points together based on patterns or characteristics. **NZ business context** Clustering helps NZ businesses segment customers, identify product affinities, or group support tickets by theme. The output is descriptive, not prescriptive. Your team must decide: "What do we do with these clusters? Which ones matter? Which drive value?" ### **Anomaly Detection** **Definition** Identifying data points or patterns that deviate significantly from the norm, often used for fraud detection, quality control, or predictive maintenance. **NZ business context** Anomaly detection flags unusual transactions, equipment failures, or customer behaviour. For NZ businesses, this is a high-value, low-risk AI use case - AI flags, humans investigate. Ask: "What's the false positive rate, and can our team handle the alert volume?" ## Business Application ### Automation **Definition** Using AI to perform repetitive tasks without human intervention, data entry, scheduling, report generation, invoice processing. **NZ business context** Automation is where AI delivers the fastest ROI, if the process is stable and well-defined. For NZ businesses, ask: "Is this process broken? If so, fix it first. AI automates the process you give it, including the inefficiencies." Automate the right thing, or you'll just scale the problem. ### Augmentation **Definition** Using AI to enhance human capabilities, providing recommendations, drafting content, surfacing insights, while humans retain decision-making authority. **NZ business context** Augmentation is the sweet spot for most NZ businesses in 2026. AI drafts, structures, flags, and suggests. Humans review, refine, approve, and decide. This approach keeps judgement in the loop while accelerating work. Ask: "What does AI prepare, and what do humans decide?" ### Orchestration **Definition** Coordinating multiple AI systems, tools, and workflows to achieve a business outcome, often involving both automation and human-in-the-loop steps. **NZ business context** Orchestration is operating logic in action. For NZ businesses, this means designing how AI and humans work together across a process, customer inquiry → AI triage → human review → AI draft response → human approval. Ask: "Who owns each step? What are the handoffs? Where does judgement happen?" ### Agentic AI **Definition** AI systems that can plan, execute multi-step tasks, use tools, and adapt based on feedback, often with minimal human intervention. **NZ business context** Agentic AI is emerging in 2026 but still experimental for most NZ businesses. These systems can book meetings, research competitors, or manage workflows autonomously. The risk: reduced visibility and control. Ask: "What's the blast radius if this agent makes a mistake? What's our override path?" ### Copilot **Definition** An AI assistant embedded in workflows to provide real-time suggestions, automate subtasks, and accelerate work (e.g., Microsoft Copilot, GitHub Copilot). **NZ business context** Copilots are augmentation tools, they draft emails, generate code, summarise meetings. For NZ businesses, the value depends on adoption and workflow integration. Ask: "Does this reduce cycle time or cognitive load? Are people actually using it? What's the cost per seat vs. productivity gain?" ### Chatbot / Conversational AI **Definition** AI systems that interact with users via text or voice, handling queries, providing information, or completing transactions. **NZ business context** Chatbots can deflect routine inquiries, freeing human teams for complex cases. For NZ businesses, the trap is deploying a chatbot that frustrates customers with rigid scripts or poor understanding. Ask: "What's the escalation path to a human? What's the resolution rate? What's the customer satisfaction score?" ### Predictive Analytics **Definition** Using historical data and machine learning to forecast future outcomes, sales, churn, demand, risk. **NZ business context** Predictive analytics helps NZ businesses anticipate rather than react. The key: predictions are probabilities, not certainties. Ask: "What's the confidence interval? What happens if the forecast is wrong? How often do we retrain the model as conditions change?" ### Prescriptive Analytics **Definition** AI that recommends specific actions to achieve desired outcomes, often combining prediction with optimisation. **NZ business context** Prescriptive analytics suggests: "Do this to reduce churn" or "Adjust pricing here to maximise margin." For NZ businesses, this is powerful, but only if recommendations are explainable and aligned with strategy. Ask: "Why is the system recommending this? Does it account for constraints we care about (brand, compliance, customer experience)?" ### Sentiment Analysis **Definition** Using NLP to determine the emotional tone of text, positive, negative, neutral, often applied to customer feedback, reviews, or social media. **NZ business context** Sentiment analysis helps NZ businesses monitor brand perception, prioritise support tickets, or identify dissatisfaction early. The trap: sarcasm, context, and cultural nuance often confuse sentiment models. Ask: "What's the accuracy rate? Are we reviewing flagged cases, or trusting the score blindly?" ### Recommendation Engine **Definition** AI systems that suggest products, content, or actions based on user behaviour, preferences, or similarity to others. **NZ business context** Recommendation engines drive e-commerce upsells, content engagement, and personalised experiences. For NZ businesses, the question is: "Does this improve customer experience and revenue, or just push inventory?" Monitor for filter bubbles and ensure recommendations align with customer value, not just algorithmic convenience. ### Optical Character Recognition (OCR) **Definition** Technology that converts images of text (scanned documents, photos, PDFs) into machine-readable text. **NZ business context** OCR unlocks data trapped in paper or image formats, invoices, contracts, forms. For NZ businesses, modern OCR (powered by AI) handles handwriting, complex layouts, and multiple languages. Ask: "What's the error rate? Do we need human review for critical documents (legal, financial)?" ### Robotic Process Automation (RPA) **Definition** Software bots that automate rule-based, repetitive tasks across applications, data entry, report generation, system updates. **NZ business context** RPA is often the first "AI" project NZ businesses attempt (though it's not true AI, it's scripted automation). RPA works well for stable, high-volume processes. The trap: RPA breaks when processes or systems change. Ask: "Is this process stable enough to automate? What's the maintenance cost? Is there a simpler integration path?" ### Computer Vision **Definition** AI that interprets and understands visual information from images or video, object detection, facial recognition, quality inspection. **NZ business context** Computer vision powers quality control, inventory management, safety monitoring, and customer analytics (foot traffic, shelf compliance). For NZ businesses, this often means using cloud APIs (Google Vision, AWS Rekognition). Ask: "What's the accuracy threshold we need? What happens when the system misidentifies something critical?" ### Speech Recognition / Speech-to-Text **Definition** AI that converts spoken language into written text, enabling voice commands, transcription, and accessibility. **NZ business context** Speech recognition powers meeting transcription, voice assistants, and call centre analytics. For NZ businesses, test for Kiwi accents, industry jargon, and background noise tolerance. Ask: "What's the transcription accuracy? Do we need human review for compliance or record-keeping?" ### Text-to-Speech (TTS) **Definition** AI that converts written text into spoken audio, often with natural-sounding voices. **NZ business context** TTS supports accessibility, voice assistants, and automated customer communications (IVR, notifications). For NZ businesses, modern TTS (like ElevenLabs, Google WaveNet) sounds increasingly human. Ask: "Does the voice match our brand? Is it clear to our audience? What's the cost per minute of audio?" ## Risk & Governance ### **Hallucination** **Definition** When an AI model generates information that is plausible-sounding but factually incorrect or entirely fabricated. **NZ business context** Hallucinations are the biggest operational risk for NZ businesses using generative AI. LLMs confidently invent facts, citations, policies, or customer details. The mitigation: never deploy generative AI without human review in customer-facing, regulated, or high-stakes contexts. Use RAG to ground outputs in verified data. No citation, no trust. ### **Bias** **Definition** Systematic errors or unfair outcomes in AI systems, often reflecting biases in training data, design choices, or societal patterns. **NZ business context** Bias can lead to discriminatory hiring, lending, pricing, or service decisions, legal and reputational risk. For NZ businesses, ask: "What data trained this model? Does it reflect our diverse customer base? Have we tested for bias across demographics (age, gender, ethnicity, location)?" Monitor outcomes, not just intentions. ### **Explainability / Interpretability** **Definition** The ability to understand and articulate how an AI system reached a specific decision or output. **NZ business context** Explainability matters for regulated industries (finance, insurance, health) and high-stakes decisions (credit, hiring, claims). For NZ businesses, ask: "Can we explain this decision to a customer, regulator, or tribunal? If not, should we be using AI here?" Black-box models may be accurate but legally or ethically risky. ### **Fairness** **Definition** Ensuring AI systems produce equitable outcomes across different groups, avoiding discrimination or disproportionate harm. **NZ business context** Fairness is both an ethical and legal obligation under NZ's Human Rights Act and Privacy Act. For NZ businesses, this means: Test AI systems for disparate impact. Monitor outcomes by demographic group. Establish fairness criteria before deployment, not after complaints arise. ### **Transparency** **Definition** Openness about how AI systems work, what data they use, and how decisions are made. **NZ business context** Transparency builds trust with customers, regulators, and employees. For NZ businesses, this means: Disclose when AI is involved in decisions. Explain how data is used. Provide avenues for questions and appeals. Transparency isn't just good ethics, it's risk management. ### **Accountability** **Definition** Clear assignment of responsibility for AI system design, deployment, outputs, and impacts. **NZ business context** When AI makes a mistake, who's accountable? For NZ businesses, this means: Assign ownership for AI systems (not just IT, business owners). Establish escalation paths. Define decision rights. AI doesn't make decisions, people do. Make sure your org chart reflects that. ### **Data Privacy** **Definition** Protecting personal information from unauthorised access, use, or disclosure, in compliance with regulations like NZ's Privacy Act. **NZ business context** AI systems often process personal data - customer details, employee records, transaction history. For NZ businesses, ask: "What data does this AI access? Where is it stored? Who can see it? Is it anonymised? Do we have consent?" Privacy breaches carry legal penalties and reputational damage. ### **Data Sovereignty** **Definition** The principle that data is subject to the laws and governance of the country where it's collected or stored. **NZ business context** Many AI platforms (OpenAI, Google, AWS) store data offshore, raising sovereignty concerns for NZ businesses, especially in regulated sectors. Ask: "Where is our data processed and stored? What jurisdiction governs it? Can we use NZ-hosted alternatives if required?" ### **Model Drift** **Definition** When an AI model's performance degrades over time because the real-world data it encounters no longer matches the training data. **NZ business context** Markets change. Customer behaviour evolves. Regulations shift. For NZ businesses, this means: Monitor model performance continuously. Retrain models when accuracy drops. Don't assume a model deployed in 2024 is still accurate in 2026. Drift is inevitable, detection and response are your responsibility. ### **Adversarial Attacks** **Definition** Intentional manipulation of AI inputs to trick the system into making incorrect predictions or outputs. **NZ business context** Adversarial attacks are rare but possible - altering images to fool computer vision, crafting prompts to bypass content filters, or injecting false data to skew predictions. For NZ businesses in high-risk domains (finance, security), ask: "What's our exposure? Do we have detection mechanisms? What's the fallback if the AI is compromised?" ### **Prompt Injection** **Definition** A type of attack where malicious users craft inputs designed to override an AI's instructions or extract sensitive information. **NZ business context** Prompt injection is a growing risk for customer-facing chatbots and AI assistants. Attackers may try to trick the AI into revealing internal data, bypassing filters, or performing unauthorised actions. For NZ businesses, mitigation includes: input validation, output filtering, and limiting AI access to sensitive systems. ### **Synthetic Data** **Definition** Artificially generated data that mimics real-world data, used for training AI models when real data is scarce, sensitive, or expensive to collect. **NZ business context** Synthetic data can accelerate AI development while protecting privacy. For NZ businesses, this is useful in regulated industries (health, finance) where real data is restricted. Ask: "Is synthetic data representative enough? Does it introduce new biases? Can we validate model performance on real data before deployment?" ### **Differential Privacy** **Definition** A mathematical framework for sharing insights from datasets while protecting individual privacy, often by adding controlled noise to data. **NZ business context** Differential privacy allows NZ businesses to use data for AI training and analytics without exposing personal information. This is critical for compliance with NZ's Privacy Act. Ask: "Does this technique meet our privacy obligations? What's the trade-off between privacy and model accuracy?" ### **Federated Learning** **Definition** A machine learning approach where models are trained across decentralised devices or servers without centralising raw data. **NZ business context** Federated learning is useful when data can't leave its source due to privacy, sovereignty, or security constraints. For NZ businesses, this enables collaboration (e.g., industry-wide fraud detection) without sharing sensitive data. Ask: "What's the coordination cost? Who manages the shared model?" ### **Red Teaming** **Definition** Deliberately testing AI systems for vulnerabilities, biases, and failure modes by simulating adversarial scenarios. **NZ business context** Red teaming helps NZ businesses identify risks before deployment. This means: testing for bias, prompt injection, hallucinations, edge cases, and adversarial inputs. Ask: "Have we tested this AI under stress? What breaks it? What's our mitigation plan for identified risks?" ## Measurement & Value ### **ROI** (Return on Investment) **Definition** The financial return generated by an AI initiative relative to its cost, typically expressed as a percentage or ratio. **NZ business context** ROI is the question every CFO asks. For NZ businesses, AI ROI must account for: implementation cost, licensing, training, maintenance, and opportunity cost. Ask: "What's the baseline? What metric are we moving? What's the timeframe? What's the cost of doing nothing?" Avoid vanity metrics. Focus on business outcomes. ### **TCO** (Total Cost of Ownership) **Definition** The full cost of acquiring, deploying, and maintaining an AI system over its lifecycle, licensing, infrastructure, training, support, updates. **NZ business context** TCO often exceeds initial estimates. For NZ businesses, hidden costs include: data preparation, integration, retraining, compliance, and change management. Ask: "What's the 3-year cost? Who pays for updates and support? What happens if the vendor raises prices or shuts down?" ### **Value Levers** **Definition** The specific business metrics AI is intended to improve - revenue, cost-to-serve, cycle time, risk, or customer experience. **NZ business context** Use cases describe what AI can do. Value levers define what the business must improve. For NZ businesses, every AI initiative must name one primary lever, a baseline, a target, and a timeframe. Ask: "Which number are you willing to be held accountable for moving?" Move a metric, not a use case. ### **Baseline** **Definition** The current state of a metric before AI intervention, used to measure impact and ROI. **NZ business context** Without a baseline, you can't measure value. For NZ businesses, this means: capture the current state before deployment, cycle time, error rate, cost per transaction, customer satisfaction. Ask: "What's the number today? How are we measuring it? Who owns the data?" ### **Proof of Concept** (POC) **Definition** A small-scale, time-boxed experiment to test whether an AI solution is feasible and valuable before full deployment. **NZ business context** POCs help NZ businesses de-risk AI investments. The trap: endless POCs without production deployment. Ask: "What's the success criteria? What's the timeline? If this works, what's the path to production? If it fails, what do we learn?" POCs should inform decisions, not delay them. ### **Pilot** **Definition** A limited deployment of an AI system with a subset of users, processes, or data to validate performance and refine before scaling. **NZ business context** Pilots bridge POC and production. For NZ businesses, this means: real users, real workflows, real stakes, but contained blast radius. Ask: "What's the pilot scope? What are we learning? What's the rollout plan if successful? What's the kill criteria if not?" ### **Scalability** **Definition** The ability of an AI system to handle increasing volume, complexity, or scope without degrading performance or requiring proportional cost increases. **NZ business context** A solution that works for 10 users may fail at 100. For NZ businesses, ask: "What's the cost curve as we scale? What's the latency at 10x volume? What infrastructure is required? Can we scale incrementally, or is this all-or-nothing?" ### **Latency** **Definition** The time delay between input and output in an AI system - how long it takes to get a response. **NZ business context** Latency matters for real-time applications, customer chatbots, fraud detection, dynamic pricing. For NZ businesses, ask: "What's acceptable response time? What's the user experience if latency spikes? What's the cost trade-off between speed and accuracy?" ### **Throughput** **Definition** The volume of tasks or requests an AI system can process in a given time period. **NZ business context** Throughput determines capacity. For NZ businesses, ask: "Can this handle peak load (month-end, holiday season, product launch)? What happens when demand exceeds capacity? Do we queue, delay, or scale infrastructure?" ### **Accuracy** **Definition** The proportion of correct predictions or outputs an AI system produces, often measured against a labelled test dataset. **NZ business context** Accuracy is necessary but not sufficient. For NZ businesses, ask: "Accurate for whom? In what context? What's the cost of a false positive vs. false negative?" A 95% accurate fraud detector sounds good, until you realise 5% of legitimate transactions are blocked. ### **Precision** **Definition** Of the items the AI flagged as positive, what proportion were actually positive? (True positives / [True positives + False positives]) **NZ business context** High precision means fewer false alarms. For NZ businesses, this matters when false positives are costly, blocking legitimate customers, flagging safe transactions, rejecting valid claims. Ask: "What's the false positive rate? Can our team handle the review volume?" ### **Recall** **Definition** Of all actual positive cases, what proportion did the AI correctly identify? (True positives / [True positives + False negatives]) **NZ business context** High recall means fewer missed cases. For NZ businesses, this matters when false negatives are dangerous, missed fraud, undetected defects, overlooked risks. Ask: "What's the false negative rate? What's the cost of missing a case?" ### **F1 Score** **Definition** A metric that balances precision and recall, useful when you need to optimise for both. **NZ business context** F1 score helps NZ businesses evaluate trade-offs. High precision but low recall? You're missing cases. High recall but low precision? You're drowning in false alarms. Ask: "What's the right balance for our use case? Which error type is more costly?" ### **Confusion Matrix** **Definition** A table showing true positives, true negatives, false positives, and false negatives, used to evaluate classification model performance. **NZ business context** Confusion matrices make trade-offs visible. For NZ businesses, this means: you can see exactly where the model succeeds and fails. Ask: "Which quadrant matters most? What's our tolerance for each error type? How do we mitigate the costly errors?" ### **Ground Truth** **Definition** The verified, accurate data used to train or evaluate AI models, the "right answer" against which predictions are measured. **NZ business context** Ground truth is only as good as your data quality. For NZ businesses, ask: "Who labels ground truth? How often is it updated? Does it reflect current reality, or is it historical?" Garbage in, garbage out. ### **Benchmark** **Definition** A standard or reference point used to evaluate AI model performance, often comparing against industry baselines or competing models. **NZ business context** Benchmarks help NZ businesses set realistic expectations. Ask: "What's the industry standard for this task? How does our model compare? Are we optimising for the right metric?" Don't chase benchmarks that don't align with business value. ## Infrastructure & Deployment ### **Cloud AI** **Definition** AI services and infrastructure hosted by cloud providers (AWS, Google Cloud, Microsoft Azure), accessed via APIs or platforms. **NZ business context** Cloud AI offers speed and scalability without infrastructure investment. For NZ businesses, trade-offs include: data sovereignty, ongoing costs, and vendor lock-in. Ask: "Where is data processed? What's the cost at scale? Can we switch providers if needed?" ### **On-Premises AI** **Definition** AI systems hosted and operated on a business's own infrastructure, rather than in the cloud. **NZ business context** On-premises AI offers control, sovereignty, and compliance benefits, critical for regulated NZ businesses (finance, health, government). Trade-offs: higher upfront cost, maintenance burden, and slower iteration. Ask: "Do we have the infrastructure and expertise? What's the 3-year TCO vs. cloud?" ### **Hybrid AI** **Definition** A deployment model combining cloud and on-premises AI, often using cloud for development and on-prem for production or sensitive workloads. **NZ business context** Hybrid approaches balance flexibility and control. For NZ businesses, this might mean: training models in the cloud, deploying on-prem for compliance; or using cloud APIs for low-risk tasks and on-prem for high-stakes decisions. Ask: "What goes where, and why?" ### **API** (Application Programming Interface) **Definition** A set of protocols that allows different software systems to communicate, often used to access AI services (e.g., OpenAI API, Google Cloud Vision API). **NZ business context** APIs let NZ businesses consume AI without building it. The trade-offs: dependency on the provider, usage-based costs, and limited customisation. Ask: "What's the cost per call? What's the rate limit? What happens if the API goes down or the provider changes terms?" ### **Model Hosting** **Definition** The infrastructure and services required to deploy and run AI models in production, handling requests, scaling, and monitoring. **NZ business context** Model hosting options include: cloud platforms (AWS SageMaker, Azure ML), third-party services (Hugging Face), or self-hosted. For NZ businesses, ask: "What's the cost? What's the latency? Who manages updates and scaling?" ### **Edge AI** **Definition** Running AI models on local devices (phones, IoT sensors, edge servers) rather than in centralised cloud infrastructure. **NZ business context** Edge AI reduces latency, bandwidth, and privacy risk, useful for real-time applications (manufacturing, retail, logistics). For NZ businesses, trade-offs include: limited compute power on edge devices and complexity of managing distributed models. Ask: "Do we need real-time response? Can we tolerate cloud latency?" ### **MLOps** (Machine Learning Operations) **Definition** Practices and tools for deploying, monitoring, and maintaining machine learning models in production, analogous to DevOps for software. **NZ business context** MLOps is the difference between a POC and a production system. For NZ businesses, this means: version control for models, automated retraining, performance monitoring, and incident response. Ask: "Who owns MLOps? What's our process for detecting and fixing model drift?" ### **CI/CD for AI** (Continuous Integration / Continuous Deployment) **Definition** Automated pipelines for testing, validating, and deploying AI models and updates, ensuring quality and speed. **NZ business context** CI/CD for AI enables rapid iteration while maintaining quality. For NZ businesses, this means: automated testing for accuracy, bias, and performance before deployment. Ask: "What's our deployment cadence? What gates must a model pass before going live?" ### **Model Registry** **Definition** A centralised repository for storing, versioning, and managing AI models, tracking metadata like training data, performance metrics, and deployment history. **NZ business context** Model registries provide governance and auditability. For NZ businesses, this means: you can track which model version is in production, roll back if needed, and audit decisions. Ask: "What's our model versioning strategy? Can we reproduce a model from 6 months ago if audited?" ### **Feature Store** **Definition** A centralised platform for managing, storing, and serving features (input variables) used by machine learning models, ensuring consistency across training and production. **NZ business context** Feature stores prevent training-serving skew (when models perform well in testing but fail in production due to inconsistent data). For NZ businesses, this is critical for operational AI. Ask: "Are we using the same features in production as in training? Who maintains the feature pipeline?" ### **Data Pipeline** **Definition** The automated workflow for collecting, cleaning, transforming, and delivering data to AI systems. **NZ business context** Data pipelines are the infrastructure behind AI. For NZ businesses, broken pipelines = broken AI. Ask: "What's our data freshness? What happens if the pipeline fails? Who monitors data quality? What's the fallback?" ### **Batch Processing** **Definition** Processing data or running AI models on large datasets at scheduled intervals (e.g., nightly, weekly) rather than in real-time. **NZ business context** Batch processing is cost-effective for non-urgent tasks, monthly reporting, customer segmentation, bulk predictions. For NZ businesses, ask: "Is real-time necessary, or can we batch? What's the cost difference? What's the acceptable delay?" ### **Real-Time Processing** **Definition** Processing data and generating AI outputs immediately as requests arrive, with minimal latency. **NZ business context** Real-time processing is essential for customer-facing applications, chatbots, fraud detection, dynamic pricing. For NZ businesses, the cost is higher (compute, infrastructure, complexity). Ask: "Do we need real-time, or is near-real-time (seconds vs. milliseconds) sufficient?" ### **A/B Testing** **Definition** Comparing two versions of an AI system (or any system) by randomly assigning users to each version and measuring outcomes. **NZ business context** A/B testing helps NZ businesses validate AI impact. For example: Does the AI-powered recommendation engine increase sales vs. the old rule-based system? Ask: "What's the test duration? What's the success metric? What's the statistical significance threshold?" ### **Shadow Mode** **Definition** Running an AI system in parallel with the existing process without affecting live operations, used to validate performance before full deployment. **NZ business context** Shadow mode de-risks deployment. For NZ businesses, this means: the AI generates outputs, but humans make decisions using the old process. You compare AI vs. human decisions to assess readiness. Ask: "How long do we shadow? What's the accuracy threshold for go-live?" ## Emerging & Advanced Concepts ### **Multimodal AI** **Definition** AI systems that process and integrate multiple types of data, text, images, audio, video, to generate outputs or make decisions. **NZ business context** Multimodal AI powers advanced applications, analysing customer service calls (voice + transcript), interpreting product images + descriptions, or generating video from text. For NZ businesses, this is emerging but increasingly accessible via APIs (GPT-4V, Gemini). Ask: "Does multimodal capability add value, or is single-mode sufficient?" ### **Foundation Model** **Definition** Large-scale AI models trained on broad datasets that can be adapted for many tasks, examples include GPT-4, Claude, Gemini, Llama. **NZ business context** Foundation models are the base layer for most generative AI applications. For NZ businesses, you don't build foundation models, you use them via APIs or fine-tune them. Ask: "Which foundation model best fits our use case? What's the cost and performance trade-off?" ### **Open Source AI** **Definition** AI models, tools, and frameworks whose code is publicly available and modifiable - examples include Llama, Mistral, Hugging Face models. **NZ business context** Open source AI offers flexibility, control, and cost savings, but requires technical expertise. For NZ businesses, ask: "Do we have the skills to deploy and maintain open source models? What's the support model? What are the licensing terms?" ### **Proprietary AI** **Definition** AI models and systems owned and controlled by vendors, typically accessed via paid APIs or licenses, examples include OpenAI's GPT-4, Anthropic's Claude. **NZ business context** Proprietary AI offers ease of use, support, and cutting-edge performance, but at the cost of dependency and ongoing fees. For NZ businesses, ask: "What's the lock-in risk? What happens if the vendor raises prices, changes terms, or shuts down?" ### **Small Language Model (SLM)** **Definition** Compact AI models optimised for efficiency, speed, and lower cost, often designed to run on-device or in resource-constrained environments. **NZ business context** SLMs (like Phi, Gemini Nano) offer privacy, speed, and cost benefits for specific tasks. For NZ businesses, ask: "Is a smaller, faster, cheaper model sufficient for this use case, or do we need the power of a large model?" ### **Mixture of Experts (MoE)** **Definition** An AI architecture that activates only relevant subsets of a model (experts) for each task, improving efficiency and performance. **NZ business context** MoE models (like GPT-4, some Mistral variants) offer large-model performance at lower computational cost. For NZ businesses using APIs, this is invisible, but it improves speed and cost. Ask: "What's the cost-performance trade-off vs. standard models?" ### **Reasoning Models** **Definition** AI systems designed to perform multi-step logical reasoning, planning, and problem-solving - examples include OpenAI's o1, o3 models. **NZ business context** Reasoning models excel at complex tasks, legal analysis, financial modelling, strategic planning. For NZ businesses, these models are slower and more expensive than standard LLMs. Ask: "Does this task require deep reasoning, or is a fast, general-purpose model sufficient?" ### **Synthetic Media / Deepfake** **Definition** AI-generated or manipulated images, audio, or video that appear authentic but are partially or entirely fabricated. **NZ business context** Synthetic media poses reputational and fraud risks, fake executive videos, manipulated evidence, impersonation scams. For NZ businesses, mitigation includes: verification protocols, watermarking, and employee awareness training. Ask: "What's our exposure? How do we verify authenticity?" ### **AI Alignment** **Definition** Ensuring AI systems behave in ways consistent with human values, intentions, and goals - a key challenge in advanced AI development. **NZ business context** Alignment matters when deploying AI with significant autonomy. For NZ businesses, ask: "Does this AI's behaviour align with our values, risk tolerance, and customer expectations? What happens when it encounters edge cases we didn't anticipate?" ### **AI Safety** **Definition** Research and practices aimed at preventing AI systems from causing unintended harm, technical failures, misuse, or unintended consequences. **NZ business context** AI safety is both a technical and governance challenge. For NZ businesses, this means: testing for failure modes, establishing guardrails, and maintaining human oversight. Ask: "What's the worst-case scenario? What's our mitigation plan?" ### **Responsible AI** **Definition** A framework for developing and deploying AI ethically, transparently, and accountably, addressing bias, fairness, privacy, and societal impact. **NZ business context** Responsible AI is both a legal obligation (under NZ's Human Rights Act, Privacy Act) and a business imperative. For NZ businesses, this means: governance frameworks, impact assessments, and ongoing monitoring. Ask: "Who owns responsible AI? What's our review process? How do we measure fairness and impact?" ## What to Do Next You don't need to memorise this glossary. You need to use it. Here's how: **Bookmark this page.** When a vendor, consultant, or colleague uses a term you don't fully understand, look it up. Interrogate the claim. **Ask better questions.** The next time someone pitches an AI solution, ask: "What's the value lever? What's the baseline? What's the human-in-the-loop workflow? What's the hallucination risk? What's the bias mitigation plan?" **Share with your team.** Send this to your leadership team, project sponsors, or AI working group. Build a shared vocabulary. Make critical thinking the norm. **Use it in procurement.** When evaluating AI vendors, use this glossary to decode marketing claims and probe technical details. "AI-powered" means nothing. "RAG-based system with explainability and human-in-the-loop approval" means something. **Teach it forward.** Curiosity is contagious. Use this glossary to help your team ask sharper questions, spot AI theatre, and focus on what moves the business. Critical thinking is your competitive moat. This glossary is your starting point. **Want help applying these concepts to your business?** Let's talk. Whether you're navigating your first AI project or scaling across the organisation, we can help you move from AI theatre to operational impact. ## Power up your potential with practical AI skills. Contact us to discuss how Artificial Intelligence could boost your business. [Book an AI Chat](https://www.newzealand.ai/contact-us/?service=ai-chat#book) [Explore the AI Operating System](https://www.newzealand.ai/ai-os/) --- # How to Choose AI Tools for Your Business: 2026 Framework Source: https://www.newzealand.ai/how-to-choose-ai-tools-for-your-business/ 3 February 2026 · Guidelines # How to Choose AI Tools for Your Business: A 2026 Comparison Framework for New Zealand Leaders This isn't a buying guide. It's a framework for making AI tool decisions that survive contact with your actual business. **Justin Flitter** · Founder, AI New Zealand · 3 February 2026 · 10 min read How to Choose AI Tools for Your Business: A 2026 Comparison Framework for New Zealand Leaders The vendor called it "AI-powered transformation." The CFO called it "another subscription we can't measure." Both were right. Across New Zealand in early 2026, business leaders face a peculiar challenge: AI tools are everywhere, vendor promises are bold, and yet most organisations still can't answer the question that actually matters - *which number will this move, and by how much?* This isn't a buying guide. It's a framework for making AI tool decisions that survive contact with your actual business - the one with tight margins, seasonal demand swings, compliance requirements, and people who already have full workloads. The question isn't "What can this AI do?" It's "What must our business improve, and is this the right tool to move that specific metric?" ### The Real Problem: Tools Before Thinking According to [Stats NZ's 2025 Digital Technologies Survey](https://www.digital.govt.nz/standards-and-guidance/strategy-and-planning/digital-capability-public-service-workforce/digital-workforce-data-and-evidence), 43% of New Zealand businesses reported experimenting with AI tools in 2024–2025. Yet fewer than 18% could demonstrate measurable business outcomes tied to those experiments. The gap isn't capability. It's clarity. Most AI tool selection processes start in the wrong place: - Vendor demos showcasing features - Peer pressure ("Our competitor just announced they're using X") - IT teams evaluating technical specifications - Executives asking "What's our AI strategy?" None of these start with the question that matters: **What specific business outcome are we trying to improve?** ### The Cost of Tool-First Thinking A mid-sized NZ manufacturing business spent $47,000 on an AI-powered inventory forecasting tool in 2025. The tool was technically excellent. It integrated with their ERP system. It produced beautiful dashboards. But their real bottleneck wasn't forecast accuracy, it was the three-day lag between forecast changes and procurement adjustments. The AI made better predictions, but the operating logic stayed the same. Forecasts still sat in email chains waiting for approval. Cycle time didn't budge. Working capital stayed locked up. The tool worked perfectly. The business outcome didn't move. That's the trap. ### The Four-Question Framework: Before You Evaluate Any Tool Before you compare features, pricing, or vendor claims, answer these four questions. If you can't answer them clearly, you're not ready to choose a tool. **Question 1: Which Value Lever Are You Moving?** AI tools don't create value in the abstract. They move specific business metrics. There are five core value levers every AI initiative should target: - **Revenue** – Increase sales, improve conversion, expand customer lifetime value - **Cost-to-serve** – Reduce operational cost per transaction, customer, or unit delivered - **Cycle time** – Compress time from trigger to outcome (quote to cash, inquiry to resolution, variation to approval) - **Risk** – Reduce error rates, compliance failures, fraud, safety incidents - **Customer experience** – Improve satisfaction, retention, NPS, effort scores Pick **one primary lever**. Name the specific metric. Define the baseline and target. **Example (good):** - Lever: Cycle time - **Metric:** Days from customer variation request to approved quote - Baseline: 4.2 days (measured Q4 2025) - Target: Under 2 days within 90 days - Owner: GM Operations **Example (bad):** - "Improve customer service with AI" - "Increase efficiency" - "Transform our operations" If you can't name the metric, you can't measure success. And if you can't measure success, you can't choose the right tool. **Question 2: What's the Real Bottleneck?** AI accelerates workflows. If the workflow is broken, AI scales the breakage. A New Zealand professional services firm implemented an AI meeting transcription tool in late 2025. Transcripts were accurate and fast. But the real bottleneck wasn't capturing what was said - it was deciding what to do with it. Meeting notes piled up. Action items stayed unclear. Accountability didn't improve. The tool worked. The operating logic didn't. **Before choosing any tool, map the current workflow:** - Where does work wait? - Where do errors happen? - Where does judgement get applied? - Where do handoffs break down? According to [Deloitte's 2025 AI Adoption Report for Asia-Pacific](https://www.deloitte.com/au/en/services/economics/analysis/ai-business-apac-trends-platform-adoption.html), 62% of failed AI implementations were attributed to "process and workflow misalignment," not technical failure. The AI didn't fail. The business wasn't ready. **Question 3: Where Does Human Judgement Stay in the Loop?** In 2026, AI boundaries are clearer than ever: - **AI drafts → humans approve** - **AI structures → humans verify** - **AI flags → humans decide** Any tool that removes human judgement from high-stakes decisions introduces risk. Any tool that buries human judgement in volume introduces fatigue. The question isn't "Can AI do this?" It's "Should AI do this, and where does human oversight happen?" **Critical thinking is your competitive moat.** AI commoditises answers. Judgement is what differentiates your business from every competitor using the same tools. When evaluating tools, ask: - What does the AI influence vs. what does it decide? - Where are override and escalation paths? - How is human judgement made visible and teachable? - What happens when the AI is confidently wrong? A [2025 Stanford HAI study](https://hai.stanford.edu/) found that professionals using AI without structured review processes were 34% more likely to accept contextually incorrect outputs - outputs that were factually accurate but strategically wrong. Not factually wrong. Contextually wrong. That's the risk. **Question 4: How Will You Measure This Weekly?** Quarterly ROI reviews are AI theatre. By the time you realise something isn't working, you've lost three months and momentum. Measurement must be: - **Weekly** (at minimum, daily for high-velocity workflows) - **Tied to the value lever** (the metric you named in Question 1) - **Owned by someone** (not "the AI team", the business owner accountable for that metric) No vanity metrics. No "number of prompts used." Just business numbers. **Example measurement plan:** - Lever: Cost-to-serve (customer support) - Metric: Average handle time per inquiry - Baseline: 8.3 minutes (Dec 2025) - Target: Under 6 minutes by March 2026 - **Tracking:** Weekly dashboard, reviewed every Monday with support lead If you can't commit to weekly tracking, the initiative isn't a priority. And if it's not a priority, don't buy the tool. The 2026 AI Tool Landscape: What's Actually Different The AI tool market in 2026 is maturing rapidly. Here's what's changed from 2023–2024: **Consolidation and Integration** Standalone point solutions are giving way to integrated suites. According to [Gartner's 2026 AI Market Forecast](https://www.gartner.com/en/newsroom/press-releases/2026-1-15-gartner-says-worldwide-ai-spending-will-total-2-point-5-trillion-dollars-in-2026), 58% of enterprise AI spending is now on platforms that combine multiple capabilities (e.g., Microsoft Copilot, Google Workspace AI, Salesforce Einstein). **What this means for NZ businesses:** - Favour tools that integrate with your existing tech stack (Microsoft 365, Google Workspace, Xero, MYOB, Salesforce) - Evaluate whether a general-purpose AI assistant (ChatGPT Enterprise, Claude for Work) meets 70%+ of your needs before buying niche tools Watch for vendor lock-in - can you export data and switch if needed? ### Regulation and Compliance New Zealand's [Privacy Act 2020](https://www.privacy.org.nz/) applies to AI-generated outputs. The [Commerce Commission's 2025 guidance on AI and consumer protection](https://comcom.govt.nz/) clarified liability for AI-driven decisions in customer-facing contexts. **What this means:** - Tools handling personal data must meet NZ privacy standards - Audit trails and explainability matter more in regulated industries (finance, health, legal) - If you're in a high-stakes domain, prioritise tools with citation and source-tracking (e.g., AI that shows its work) ### Cost Structures Are Shifting Pricing models in 2026 range from per-seat subscriptions to usage-based (per API call, per token, per transaction). A [PwC New Zealand survey](https://www.pwc.co.nz/) found that 41% of businesses underestimated AI tool costs by more than 30% in their first year due to usage-based pricing surprises. **What this means:** - Model realistic usage before committing - Factor in training, integration, and change management costs (often 2–3× the license cost) - Start small, measure, scale - don't over-commit upfront - Tool Comparison Framework: Evaluating Your Shortlist Once you've answered the four questions, you're ready to evaluate specific tools. Use this framework to compare options. **Capability Fit (40% weighting)** - Does it directly address the value lever you named? - Does it integrate with your current systems (CRM, ERP, email, Slack, etc.)? - Can it handle NZ-specific context (spelling, regulations, time zones, business norms)? **Operating Logic Fit (30% weighting)** - Where does human judgement stay in the loop? - Can you configure approval, override, and escalation workflows? - Does it make decisions transparent and auditable? **Measurement and Control (20% weighting)** - Can you track the metric you care about weekly? - Does it provide usage data, error rates, and outcome tracking? - Can you export data for independent analysis? **Cost and Risk (10% weighting)** - What's the total cost (license + integration + training + ongoing management)? - What's the exit cost if it doesn't work? - Does the vendor have NZ presence or support in NZ-friendly time zones? ### Common AI Tool Categories and When to Use Them Here's a practical breakdown of common AI tool categories, mapped to value levers and use cases relevant to New Zealand businesses. **Generative AI Assistants** (ChatGPT, Claude, Gemini, Copilot) **Best for:** - Drafting content (emails, reports, proposals, policies) - Structuring unstructured information (meeting notes, research, customer feedback) - Rapid prototyping (business cases, process maps, brainstorming) **Value levers:** Cycle time, cost-to-serve **NZ context:** Works well for professional services, marketing, HR, legal drafting. Requires strong critical thinking to catch contextual errors. **Watch out for:** Over-reliance without review; confidently wrong outputs in specialised domains. **Customer Service AI** (Zendesk AI, Intercom, Ada, Freshdesk AI) **Best for:** - Handling high-volume, repetitive inquiries - Routing and triaging support tickets - Providing 24/7 first-response capability **Value levers:** Cost-to-serve, customer experience, cycle time **NZ context:** Effective for e-commerce, SaaS, utilities, telcos. Ensure escalation to humans for complex or sensitive issues. **Watch out for:** Generic responses that frustrate customers; lack of NZ-specific knowledge (public holidays, regional nuances). **Sales and CRM AI** (Salesforce Einstein, HubSpot AI, Pipedrive AI) **Best for:** - Lead scoring and prioritisation - Email automation and follow-up sequencing - Forecasting and pipeline analysis **Value levers:** Revenue, cycle time **NZ context:** Useful for B2B sales teams, real estate, recruitment. Works best when CRM data is clean and current. **Watch out for:** Garbage in, garbage out - AI can't fix poor data hygiene. **Document and Knowledge AI** (Notion AI, Glean, Guru, SharePoint AI) **Best for:** - Searching internal knowledge bases - Summarising long documents (contracts, reports, research) - Answering employee questions from company content **Value levers:** Cycle time, cost-to-serve (internal) **NZ context:** High value for organisations with dispersed teams, complex compliance docs, or high onboarding costs. **Watch out for:** Requires well-organised, up-to-date source content; hallucination risk if sources are incomplete. **Process Automation AI** (UiPath AI, Automation Anywhere, Microsoft Power Automate AI) **Best for:** - Automating repetitive data entry, reconciliation, reporting - Extracting information from invoices, forms, emails - Triggering workflows based on conditions **Value levers:** Cost-to-serve, cycle time, risk (error reduction) **NZ context:** Strong fit for finance, logistics, government services, insurance. ROI often clear and measurable. **Watch out for:** Brittle if underlying processes change frequently; requires IT support for setup and maintenance. **Industry-Specific AI** (legal AI, medical AI, accounting AI, etc.) **Best for:** - Specialised tasks requiring domain knowledge (contract review, diagnostic support, tax compliance) **Value levers:** Risk, cycle time, cost-to-serve **NZ context:** Evaluate NZ regulatory alignment carefully (e.g., legal AI trained on US law may not apply). Favour NZ vendors or those with local partnerships. **Watch out for:** High cost, long integration timelines, vendor lock-in. Your 7-Day Action Plan: Choosing the Right AI Tool Here's what to do this week to move from vendor demos to real decisions: **Pick one value lever.** Choose revenue, cost-to-serve, cycle time, risk, or customer experience. Name the specific metric you'll move. **Define baseline and target.** Measure where you are now. Set a 90-day target that's ambitious but realistic. **Map the current workflow.** Draw out the steps, handoffs, wait times, and decision points. Identify the real bottleneck. **Clarify human-in-the-loop boundaries.** Decide what AI influences, what requires review, and what stays human-only. **Shortlist 2–3 tools.** Use the comparison framework (capability fit, operating logic fit, measurement, cost/risk). Favour tools that integrate with your existing stack. **Run a 30-day pilot.** Test with a small team on a real workflow. Track the metric weekly. Adjust or kill fast. **Scale or stop.** If the metric moves, scale. If it doesn't, stop and reassess. Don't let sunk cost drive bad decisions. ### The Bottom Line: Tools Don't Create Value - Decisions Do Over the next 12 months, New Zealand businesses will spend millions on AI tools. Some will move metrics. Most will sit unused or underutilised. The difference won't be the tool. It'll be the thinking that came before the purchase. AI doesn't fix broken processes. It accelerates them. AI doesn't replace judgement. It commoditises answers and makes human judgement premium. The businesses that win in 2026 won't be the ones with the most AI tools. They'll be the ones that ask better questions: - Which number are we moving? - Where's the real bottleneck? - Where does human judgement stay in the loop? - How will we measure this weekly? Answer those questions first. Then choose the tool. And if you can't answer them, you're not ready to buy anything. **"AI use cases" are a distraction. Here's what we did instead.** **From AI Theatre to Real Impact: A NZ Financial Services Story.** Contact us to discuss how Artificial Intelligence could boost your business. ## Power up your potential with practical AI skills. Contact us to discuss how Artificial Intelligence could boost your business. [Book an AI Chat](https://www.newzealand.ai/contact-us/?service=ai-chat#book) [Explore the AI Operating System](https://www.newzealand.ai/ai-os/)