SOVATA
Last updated 23 July 2026
The AI Advisory Pilot is a New Zealand Government programme, delivered through the Regional Business Partner Network and backed by MBIE, that co-funds up to 50% (capped at $15,000) of the cost for a small or medium business to work with an approved adviser and produce a tailored AI plan. It is designed to help Kiwi businesses adopt AI safely and practically — this guide explains how it works, who qualifies, how to apply, and how to turn a funded plan into results.
Quick summary: the AI Advisory Pilot at a glance
If you only read one section, read this. The AI Advisory Pilot lowers the cost and the risk of your first serious step into AI by putting an experienced adviser alongside you and covering up to half the bill. It funds thinking and planning — a clear, tailored roadmap for your business — rather than software licences or hardware.
- What it is: government co-funding for a professional AI advisory engagement that produces a tailored AI plan for your business.
- Who runs it: MBIE, delivered through the Regional Business Partner Network (RBPN) — the same regional providers that already connect thousands of Kiwi businesses to growth support.
- How much: co-funding of up to 50%, capped at $15,000, toward the cost of your AI plan. You contribute the balance.
- Who it is for: New Zealand small and medium businesses that meet the eligibility criteria (covered below).
- What it produces: a documented AI plan aligned to your operations, your staff and your customers — not a generic report.
- How to start: contact your local Regional Business Partner to check eligibility and be matched with an approved AI adviser.
Key takeaways
- The AI Advisory Pilot funds a plan, not a purchase — the deliverable is a tailored roadmap you own and can act on.
- Co-funding covers up to 50% of the engagement, to a maximum government contribution of $15,000, so a $30,000 engagement could cost your business as little as $15,000.
- Funding is administered independently through the Regional Business Partner Network; advisers do not approve their own clients’ funding.
- The programme is deliberately practical — the goal is AI you can safely use in day-to-day operations like finance, HR, customer service and administration.
- A funded plan is only as valuable as the implementation that follows. Businesses that prepare their data, governance and people get far more from it.
- Where you deploy AI matters as much as whether you do — Private AI, data sovereignty and governance protect you as you scale.
Government announcement overview: what MBIE announced
In January 2026 the New Zealand Government announced the AI Advisory Pilot, a programme to give small businesses hands-on, practical support to adopt artificial intelligence. Small Business Minister Chris Penk framed it around a common problem: business owners want to use AI to free up time, but many are unsure where to start or how to use these tools safely. The pilot was allocated $765,000 from within existing MBIE appropriations and delivered through the established Regional Business Partner Network rather than a new bureaucracy.
Following strong demand, the Government expanded the pilot in 2026 — widening eligibility, increasing the number of businesses it can support to up to 150, and extending the programme’s run so more firms can take part. The direction of travel is clear: this is a test-and-learn programme the Government is actively scaling, not a one-off announcement. Because the details can change, always confirm the current figures and dates with the official sources listed near the end of this guide.
What is the AI Advisory Pilot?
The AI Advisory Pilot is a co-funding programme that pairs a New Zealand small or medium business with an approved AI adviser to produce a tailored AI plan. The Government contributes up to 50% of the cost (to a maximum of $15,000) and the business pays the rest. It is delivered through the Regional Business Partner Network and is designed to make the first, hardest step into AI — working out what to do and how to do it safely — cheaper and less risky.
Crucially, the pilot funds advice and planning, not tools. The output is a documented plan: an assessment of where AI could genuinely help your business, which opportunities to pursue first, what data and systems are involved, and what to watch for around privacy, security and responsible use. Think of it as buying clarity and a credible roadmap before you spend money on implementation — the stage where poorly planned AI projects most often go wrong.
Why was the AI Advisory Pilot introduced?
The programme exists because there is a gap between AI’s potential for New Zealand and the reality that most small businesses do not know where to begin. AI adoption is uneven: many firms experiment with public chatbots, but few have a considered plan for using AI safely across their operations. The pilot is a targeted intervention to close that gap — giving SMEs access to credible, independent advice so they can move from ad-hoc experimentation to deliberate, productive use.
For the wider economy to capture that upside, adoption has to reach the small and medium businesses that make up the overwhelming majority of New Zealand firms — not just large enterprises with in-house expertise. That is the strategic logic behind the pilot: lower the barrier for SMEs, and productivity gains compound across the economy.
How much funding is available through the AI Advisory Pilot?
The AI Advisory Pilot co-funds up to 50% of the cost of an approved AI advisory engagement, capped at a government contribution of $15,000 per business. Your business pays the remaining share. So a $10,000 engagement could attract up to $5,000 in co-funding; a $30,000 engagement could attract the full $15,000 maximum. The funding applies to the advisory and planning work, not to software subscriptions, hardware or the cost of building the systems afterwards.
It helps to think of the funding as a subsidy on clarity. The most expensive AI mistakes are made in the dark — buying tools that do not fit, entering sensitive data into services that were never assessed, or launching projects with no measure of success. Co-funded advice front-loads the thinking so the money you spend on implementation later is spent well.
Funding example table: what could co-funding look like?
The examples below are illustrative, to show how the 50% / $15,000 mechanics work in practice. They are not quotes, and actual engagement costs vary by scope and provider. The co-funding shown is the maximum available at 50%, up to the $15,000 cap.
| Engagement cost | Possible co-funding (up to 50%) | Your business contribution |
|---|---|---|
| $6,000 | Up to $3,000 | From $3,000 |
| $10,000 | Up to $5,000 | From $5,000 |
| $20,000 | Up to $10,000 | From $10,000 |
| $30,000 | Up to $15,000 (cap reached) | From $15,000 |
| $40,000 | Up to $15,000 (cap reached) | From $25,000 |
Notice what happens past $30,000: because the government contribution is capped at $15,000, larger engagements simply mean your business funds a bigger share above the cap. For most SMEs a well-scoped advisory engagement sits comfortably below that ceiling, which is why scoping the work sensibly matters.
Who may benefit from the AI Advisory Pilot?
The pilot is aimed at established New Zealand small and medium businesses that are ready to use AI but need expert help to do it well. Reported eligibility centres on being a genuine, trading New Zealand business of SME size, with your compliance obligations in order, and able to co-fund your share of the engagement. Final eligibility is confirmed by your Regional Business Partner provider — and because criteria have already been widened once, always check the current requirements before you apply.
- A New Zealand–registered business with a clear New Zealand footprint.
- A small or medium business — typically fewer than around 250 staff (thresholds are set by the programme and have been broadened; confirm the current figure).
- Trading for a reasonable period — reported criteria reference operating for at least 12 months.
- No outstanding compliance issues with agencies such as Inland Revenue (IRD) or WorkSafe.
- Able to co-fund the portion of the engagement the Government does not cover.
- Not already receiving other government funding for the same AI work.
Beyond the formal criteria, the businesses that get the most from the pilot share a mindset: they have a real problem they want AI to help with, a willingness to change how they work, and someone internally who will own the outcome. AI advice landing on a desk with no one accountable for acting on it is the most common way funded potential is wasted.
What AI projects could be suitable for funding?
The pilot is deliberately practical, focused on AI that helps with the everyday running of a business rather than speculative moonshots. An AI plan typically identifies a handful of high-value, lower-risk opportunities to start with. Common candidates include:
- Customer service: drafting responses, answering routine enquiries after hours, and triaging messages so staff handle the ones that need judgement.
- Administration and operations: summarising documents, extracting information from forms and emails, and cutting repetitive digital busywork.
- Finance and back office: reconciling and categorising transactions, drafting reports, and speeding up invoice and expense processing with human review.
- HR and people: helping draft policies and position descriptions, summarising documentation, and answering common staff questions from approved internal sources.
- Sales and marketing: producing first drafts of content, tailoring proposals, and analysing customer feedback at scale.
- Knowledge and search: making your own policies, procedures and records findable through natural-language questions, grounded in your approved documents.
A good adviser will pressure-test each idea against three questions: is the value real and measurable, is the data involved safe to use this way, and can a person stay accountable for the output? Opportunities that pass all three are where funded plans deliver.
How do businesses begin?
You begin with the Regional Business Partner Network, not with a software vendor. Your local Regional Business Partner is the front door: they check your eligibility, help you access the co-funding, and connect you with an approved AI adviser suited to your situation. The process is designed to be light-touch and practical.
- Step 1 — Get in touch with your local Regional Business Partner (find your regional provider via business.govt.nz).
- Step 2 — Talk through your business, your goals for AI, and confirm your eligibility for the pilot.
- Step 3 — Get matched with an approved AI adviser and agree the scope and cost of the engagement.
- Step 4 — Confirm the co-funding so you know your contribution before work begins.
- Step 5 — Work with the adviser through discovery and planning to produce your tailored AI plan.
- Step 6 — Move from plan to implementation — the stage the funding sets you up for, but does not itself deliver.
A realistic timeline runs from an initial conversation with your Regional Business Partner, to matching and scoping over a week or two, to an advisory engagement measured in weeks rather than months. The plan is the milestone that ends the funded stage; implementation is the work that turns it into results.
Why should New Zealand businesses consider AI now?
AI has moved from novelty to normal business tooling faster than almost any technology before it, and the gap between businesses that use it well and those that ignore it is widening. Adopting AI now — carefully — means reclaiming hours lost to repetitive work, responding to customers faster, and giving small teams capacity that used to require more headcount. Doing nothing is itself a decision, and increasingly a costly one as competitors compound small efficiency gains.
The important word is carefully. The businesses that win with AI are not the ones that adopt fastest; they are the ones that adopt deliberately — with a plan, with their data protected, and with clear ownership. That is precisely what the AI Advisory Pilot is designed to fund. Government co-funding lowers the cost of doing this properly, which makes now a genuinely good time to start.
Public AI vs Private AI: what’s the difference?
Public AI tools run on shared, vendor-operated infrastructure — you send data to them and receive a response. Private AI runs inside infrastructure you control, so your data is processed within a boundary you define, under your governance rules. For many everyday, low-sensitivity tasks a well-configured public tool is fine. For regulated, client-confidential or commercially sensitive information, Private AI is what lets you use AI without losing control of your data.
| Consideration | Public AI tools | Private AI |
|---|---|---|
| Where data is processed | On the vendor’s shared, often offshore, infrastructure | Inside infrastructure you control, in a location you choose |
| Data sovereignty | Hard to guarantee data stays in New Zealand | Data can be kept onshore and under NZ control |
| Control over updates | Vendor updates the model on their schedule | Updates applied deliberately, tested, on your schedule |
| Governance & audit | Limited visibility into how data is handled | Full access controls and logging you own |
| Cost shape | Low to start, usage-metered and harder to predict | Higher upfront design, more predictable running cost |
| Best for | Everyday, low-sensitivity tasks | Regulated, confidential or sovereignty-sensitive work |
A funded AI plan should tell you honestly which tasks are safe for public tools and which need a private, governed approach. That judgement — matching the deployment model to the sensitivity of the data — is one of the most valuable outputs of good advice.
DIY AI vs professional AI implementation
Plenty of businesses start with AI on their own, and for simple productivity gains that is fine. The limits show up when AI touches real business processes and real data: unmanaged tools spread, sensitive information ends up in places no one assessed, and pilots stall because no one owns the governance. Professional implementation exists to avoid those failure modes — not to gatekeep the technology, but to make it safe to rely on.
| Factor | DIY AI | Professional implementation |
|---|---|---|
| Speed to first result | Fast for simple tasks | Slightly slower, but built to last |
| Data protection | Depends on staff choices, often unmanaged | Assessed, governed and controlled by design |
| Governance & compliance | Usually informal or absent | Aligned to Privacy Act 2020 and your obligations |
| Scalability | Breaks down as use grows | Designed to scale across teams and workflows |
| Ownership & accountability | Often unclear | Defined owners and audit trails |
| Best for | Individual productivity, low-risk tasks | Business-critical, data-sensitive or regulated use |
The AI Advisory Pilot effectively subsidises the move from DIY to a professional footing. It puts an experienced adviser alongside you at the planning stage, so the systems you build afterwards are governed from the start rather than retrofitted once something goes wrong.
How does AI benefit different industries?
AI is general-purpose, but the highest-value use cases differ by sector. The table below shows practical, lower-risk starting points across industries common in the New Zealand economy — the kinds of opportunities a funded AI plan often surfaces first.
| Industry | High-value AI use cases | Key consideration |
|---|---|---|
| Professional services | Drafting, research, document review and summarisation | Client confidentiality — often points to a private, governed deployment |
| Healthcare | Administrative load, note summarisation, patient enquiries | Health information privacy and human accountability for decisions |
| Trades & construction | Quoting, scheduling, compliance paperwork, RFI responses | Accuracy and keeping a person accountable for commitments |
| Retail & e-commerce | Customer support, product content, demand and feedback analysis | Customer data handling and brand consistency |
| Manufacturing | Documentation, maintenance records, vendor and process queries | Integrating AI with existing operational systems |
| Financial & advisory | Reconciliation, reporting drafts, transaction categorisation | Regulatory obligations and data sensitivity |
How should you prepare your business for AI?
The best time to prepare is before your advisory engagement, so the funded time is spent on strategy rather than basics. Preparation is mostly organisational, not technical: knowing your priorities, understanding your data, and deciding who owns the outcome. A little groundwork dramatically improves the quality of the plan you get.
- Name the problems first: list the tasks that eat your team’s time or frustrate customers — AI should target these, not the other way around.
- Map your data: know what information you hold, how sensitive it is, and where it lives, so advisers can judge what is safe to use.
- Assign an owner: nominate someone internally accountable for AI adoption, with the authority to act on the plan.
- Set a simple success measure: decide, up front, what a good outcome looks like (time saved, faster responses, fewer errors).
- Check your obligations: understand your responsibilities under the Privacy Act 2020 and any sector-specific rules before AI touches personal data.
- Bring your team along: adoption succeeds when staff understand why it is happening and how it helps them, not when it is imposed.
What are the most common AI adoption mistakes?
Most AI disappointment traces back to a small set of avoidable mistakes — nearly all of them made before any technology is chosen. Knowing them in advance is half the protection.
- Starting with a tool instead of a problem — buying AI because it is available, not because it solves something specific.
- Ignoring data sensitivity — entering confidential or regulated information into public tools that were never assessed for it.
- No governance — running AI with no policy on what staff may do, and no accountability for outputs.
- No owner — a plan with no one responsible for acting on it, so momentum dies after the report is delivered.
- Skipping the people — deploying AI without training or buy-in, so staff quietly ignore it.
- No measure of success — never defining what good looks like, so you can’t tell whether it worked.
- Treating a plan as the finish line — a funded AI plan is a starting point; the value is in the implementation that follows.
Why does Private AI matter for responsible AI adoption?
As AI moves from a novelty to a core part of how your business runs, the question shifts from “can we use AI?” to “can we trust how we’re using it?” Responsible adoption rests on four foundations: data sovereignty (knowing where your data is processed and keeping sensitive information onshore), AI governance (clear rules and accountability for how AI is used), secure deployment (access controls, logging and an audit trail), and human accountability (a person answerable for decisions that affect people). Private AI is what makes these foundations achievable in practice, because it puts the model, the data pipeline and the controls inside a boundary you define.
This is why the deployment model is not an afterthought. AI agents that take actions on your behalf, automations that touch customer records, and assistants grounded in your internal knowledge all handle information you are obligated to protect. A private, governed deployment lets you get the productivity of AI without exporting your data — and without inheriting the risk that a public tool changes, is breached, or trains on inputs you never intended to share. Getting this right at the planning stage, while advice is co-funded, is far cheaper than remediating it later.
How Sovata.ai helps New Zealand businesses adopt AI
Sovata.ai is a New Zealand Private AI implementation partner. We help businesses move from a plan to a working, governed AI system that keeps your data under your control. Our engagements are staged so you can make confident decisions at each step, and our discovery conversations are free precisely so we can tell you honestly when a simpler, cheaper approach is enough. What we help businesses with:
- AI Discovery Sessions — a free, no-obligation conversation to understand your goals and where AI could genuinely help.
- AI Readiness Assessments — an honest read on how ready your data, people and processes are for AI.
- AI Strategy Workshops — a focused, fixed-scope session that turns intent into a prioritised, practical plan.
- Private AI Deployment — AI systems that run inside infrastructure you control, keeping sensitive data onshore.
- AI Agents — assistants that safely take on defined tasks within your rules and access controls.
- AI Automation — removing repetitive digital work from your team’s day, with human oversight where it matters.
- AI Integration — connecting AI to the systems and knowledge your business already runs on.
- AI Governance — the policies, controls and audit trails that make AI safe to rely on.
- Ongoing Support — helping you operate, measure and evolve your AI as your needs grow.
An important clarification: Sovata.ai does not administer or approve the AI Advisory Pilot. Funding decisions are made independently through the Regional Business Partner Network. We are not the programme, and we cannot guarantee, approve or improve your chances of funding. What we can do is help you make the most of a funded plan — or help you adopt AI well whether or not you use the pilot at all. If an advisory engagement is the right first step for you, your Regional Business Partner can connect you with an approved adviser.
Official government resources
For the authoritative, current details on the AI Advisory Pilot — including eligibility, funding amounts and dates, which can change over time — always rely on official New Zealand Government sources:
- MBIE — “More practical AI support for small businesses”: www.mbie.govt.nz/about/news/more-practical-ai-support-for-small-businesses
- MBIE — “AI-powered tools to support small business growth”: www.mbie.govt.nz/about/news
- business.govt.nz — practical AI guidance and your local Regional Business Partner: www.business.govt.nz
- Regional Business Partner Network — the front door to check eligibility and access co-funding (via business.govt.nz).
Book an AI Discovery Session
If you are considering the AI Advisory Pilot, or simply want a clear, honest read on where AI fits in your business, the best first step is a conversation. A free AI Discovery Session with Sovata.ai helps you understand your options — including whether the pilot is right for you — before you commit to anything. Educate first, decide second: you should finish that conversation knowing more, whether or not you ever work with us. You can book a discovery session, run our free AI readiness assessment, or explore our free AI tools to get started.
Frequently asked questions
No. It is co-funding applied to the cost of an approved AI advisory engagement — the Government covers up to 50% (capped at $15,000) of the cost of producing your AI plan, and you pay the balance. It is not a cash grant paid to your business, and it funds advice and planning rather than software, hardware or implementation.
You start with your local Regional Business Partner, which you can find through business.govt.nz. They confirm your eligibility, help you access the co-funding, and match you with an approved AI adviser. You do not apply to Sovata.ai or to any private adviser for the funding itself — funding is administered through the Regional Business Partner Network.
No. Sovata.ai does not administer or approve the AI Advisory Pilot, and cannot guarantee or improve funding approval. Funding decisions are made independently through the Regional Business Partner Network. Sovata.ai is a Private AI implementation partner that can help you make the most of a funded plan, or adopt AI well independently of the pilot.
The plan is the starting point, not the finish line. After it is delivered, the value comes from implementation — building the AI systems it recommends, with the right governance and data controls. Businesses that treat the plan as a springboard into a well-governed deployment get far more from the funding than those that file it away.
Often yes — many eligible businesses already use public AI tools informally. A funded plan is frequently used to bring that ad-hoc use under control: assessing what data is safe to use, closing governance gaps, and deciding which tasks need a private, governed approach. Confirm your specific eligibility with your Regional Business Partner.
Possibly. The pilot is a test-and-learn programme that has already been expanded once — with wider eligibility, more businesses supported and an extended run. Because figures and criteria can change, always confirm the current details with MBIE, business.govt.nz or your Regional Business Partner before applying. This guide notes its own last-updated date for the same reason.
Disclaimer and last updated
This guide is provided for general information only and is not legal, financial or professional advice. Sovata.ai does not administer, approve or influence the AI Advisory Pilot; funding decisions are made independently through the Regional Business Partner Network. Programme details — including funding amounts, eligibility criteria and dates — are set by the New Zealand Government and may change over time. Always confirm current details with MBIE, business.govt.nz or your Regional Business Partner before making decisions. This article is maintained and its last-updated date is shown near the top of the page; the programme information here was accurate to the best of our knowledge as at 23 July 2026.