Artificial intelligence is not a coming revolution; it is happening right now, and the gap between organisations that have embraced it and those that have not is widening every quarter. For New Zealand, a country that prides itself on ingenuity and punching above its weight, the data on AI adoption is a wake-up call that deserves serious attention.
This article examines exactly where New Zealand stands, why the gap exists, what it is costing the economy, where genuine progress is being made, and what businesses can do today to catch up and get ahead.
The data: New Zealand's AI gap in numbers
Before diagnosing causes, it helps to understand the scale of the challenge. Several credible data sources paint a consistent picture of a country that is behind the curve: not dramatically, but meaningfully, and in ways that compound over time.
Last OECD country to adopt a national AI strategy
In July 2025, New Zealand released its first-ever national AI strategy, titled Investing with Confidence. The timing made New Zealand the final OECD member to produce such a framework. Every comparable nation, from Singapore and Estonia to Norway and Portugal, had already done so. This was not merely a symbolic delay. Without a national strategy, businesses and public agencies had less policy certainty, less access to coordinated public investment, and no shared language for responsible AI governance.
The strategy arrived with ambitious numbers: AI is projected to add NZ$76 billion to the economy by 2038. But realising that figure depends entirely on whether adoption accelerates at a pace the current baseline does not guarantee.
SME adoption: the stark gap with Australia
The most striking data point in New Zealand's AI story is the contrast with Australia among small and medium-sized enterprises. According to the Ministry of Business, Innovation and Employment, 68% of New Zealand SMEs reported no plans to evaluate or invest in AI, compared with just 38% of Australian SMEs holding the same position. That is a 30-percentage-point gap between two neighbours who share an economic outlook, a language, and broadly similar business conditions.
SMEs are not a peripheral part of the New Zealand economy. They represent the overwhelming majority of registered businesses and a substantial share of employment. If the SME sector sits out AI adoption, the productivity gains that economists are projecting simply cannot materialise at scale.
| Indicator | New Zealand | Australia | Singapore |
|---|---|---|---|
| SMEs with no AI plans | 68% | 38% | ~20% |
| Overall AI adoption rate | 37.6% | ~52% | 70%+ |
| Large org AI usage (2024) | 67% | ~75% | ~85% |
| Workers with AI training | 24% | ~40% | ~60% |
| National AI strategy published | July 2025 | 2023 | 2019 |
Sources: MBIE, Microsoft AI Diffusion Report 2025, NZ AI Forum, IMDA Singapore. Approximate figures for illustrative comparison.
Where New Zealand stands globally
Microsoft's 2025 AI Diffusion Report placed New Zealand's overall AI adoption rate at 37.6%. That figure puts Aotearoa well behind Singapore (70%+), and trailing most comparable small advanced economies. The MBIE's own assessment of New Zealand's state of play acknowledges that when compared with similarly sized nations (Singapore, Norway, Finland, Estonia, and Portugal), New Zealand trails on most measures: research environment, start-up ecosystem, skills pipeline, and deployment scale. The gap is not irreversible, but closing it requires more than optimism.
Why New Zealand has fallen behind
New Zealand's AI lag is not the result of a single failure. It reflects a convergence of structural, cultural, and resource-related factors that have compounded over several years. Understanding them is necessary before businesses can address them.
The skills deficit
Skills are the most cited barrier to AI adoption, and the numbers are stark. According to the NZ AI Index 2025, 32% of organisations cite lack of internal capability and skills as the primary blocker to scaling AI. The 2024 Datacom State of AI Index found that 43% of non-adopters specifically named the absence of in-house expertise as their main reason for not starting. Meanwhile, 76% of New Zealand workers have received no AI training at all.
The problem operates at multiple layers simultaneously. At the technical end, there are too few data scientists and AI engineers to meet growing demand. In the middle of organisations, managers lack the skills to identify where AI creates value or how to integrate it into existing workflows. At the executive level, boards and leadership teams often lack the strategic literacy needed to commit to AI investment with confidence. And on the front line, workers are left to navigate AI tools with no formal guidance, creating risks as well as missed productivity.
Only 23% of New Zealand organisations currently employ AI specialists, and just 11% plan to hire them within the next year. The skills gap is not closing fast enough on its own.
Not sure where your team's AI capability gaps are?
We run structured AI readiness assessments that identify the exact skills and process gaps holding your organisation back.
SME hesitancy and resource constraints
New Zealand's economy is dominated by small businesses, and small businesses face genuine barriers that large enterprises can absorb more easily. Thirty-four percent of NZ businesses cite financial constraints as a barrier to AI adoption. For an owner-operator running a 12-person firm, the upfront cost of evaluating, purchasing, integrating, and training for an AI solution can feel prohibitive, particularly when the ROI is not yet obvious and the vendor landscape is crowded and confusing.
This is compounded by a scarcity of time. AI implementation done properly requires discovery work, stakeholder alignment, process mapping, piloting, and iteration. Many SME owners are already stretched across operations, sales, compliance, and HR. Dedicating bandwidth to AI evaluation simply does not feel like a priority when there are more pressing fires to manage.
The result is a rational but costly deferral: "We'll get to AI later." But as larger competitors and offshore players accelerate their adoption, the window for catching up narrows each quarter.
Governance gaps and shadow AI risk
A telling data point from the NZ AI Index: only 55% of New Zealand organisations have clear internal AI policies, yet over 50% are already experiencing "shadow AI": employees using AI tools that have not been approved, reviewed, or integrated into any governance framework. This is a dangerous combination. Staff are using tools like ChatGPT, Copilot, and various AI writing and analysis platforms to do their jobs, but without guardrails, data handling policies, or quality controls in place.
Shadow AI creates real risk: sensitive data can be exposed to third-party AI systems, outputs can be used without verification, and organisations lose the ability to audit or understand how decisions are being made. The fact that most organisations have not locked down their AI governance before widespread usage occurs means they are building risk exposure without realising it.
Only 28% of NZ organisations conduct employee training on AI tools. The combination of widespread informal use and minimal structured training is a governance problem waiting to become a liability.
The culture of caution
New Zealand has historically been cautious about adopting new technologies at scale, preferring to watch early-adopter markets absorb the growing pains before committing. That strategy has served some purposes in the past (avoiding costly enterprise software mistakes, for example), but AI is not behaving like a conventional technology cycle. The pace of capability improvement, the network effects of early adoption, and the compounding productivity advantages mean that a wait-and-see posture carries a cost that is growing every month.
Only 44% of New Zealanders believe the benefits of AI outweigh the risks. Public scepticism has translated into business hesitancy, which has translated into slower investment. That sentiment is shifting as real-world results become visible, but shifting it faster requires leadership from business owners, government, and industry bodies alike.
The cost of standing still
The data on what New Zealand stands to gain and lose from its current trajectory should sharpen the urgency of the conversation considerably.
The $3.4 billion near-term opportunity
Microsoft's 2025 New Zealand AI Economy Report identified a NZ$3.4 billion near-term opportunity across productivity, innovation, and new economic activity. This is not speculative future value; it represents realisable gains from technologies that exist today, deployed in ways that are already working in comparable economies. The question is not whether the value is there, but whether New Zealand businesses will capture it or watch it accrue elsewhere.
Over the longer horizon, the government's own strategy projects that AI could add NZ$76 billion to the economy by 2038. That is a number large enough to materially shift New Zealand's per-capita GDP trajectory. But every year of delayed adoption is a year of compound interest lost on that projection.
Productivity left on the floor
Among New Zealand workers who are using AI, the results are striking: 93–96% report increased efficiency. Those using AI for core tasks are processing more work, making fewer errors, and spending more time on higher-value activities. Scaled across the workforce, even modest productivity gains translate into significant economic output.
The productivity paradox in New Zealand is acute: the tools exist, the evidence of benefit is overwhelming, and yet the majority of workers and businesses are not using them. In a country that has wrestled with persistent productivity challenges relative to comparable OECD nations, AI represents one of the most accessible levers available, and it is largely unused.
Competitive risk as peers accelerate
The competitive risk is not just about technology; it is about talent, investment, and trade. Businesses in Australia, Singapore, and further afield that have invested early in AI capability are now operating with lower cost structures, faster decision cycles, and more responsive customer experiences. When those businesses compete in markets where New Zealand firms also operate (financial services, agriculture technology, professional services, or manufacturing), the AI-enabled competitor has structural advantages that compound over time.
For New Zealand exporters, the risk is compounded. Competing in global markets against organisations that have automated significant portions of their operations and decision-making is already harder than it was two years ago. It will be harder still two years from now if the gap continues to widen.
Green shoots: where New Zealand is catching up
The picture is not uniformly bleak. There are genuine reasons for optimism, and the trajectory among leading New Zealand organisations suggests that the capability to catch up exists; it simply needs to be accelerated and broadened.
Large enterprise adoption is rising fast
Among large New Zealand organisations, AI adoption has grown sharply: from 48% in 2023 to 67% in 2024, and now reaching 87% according to the NZ AI Index 2025. That is a substantial acceleration. Of those using AI, 88% report a positive impact, and 89% cite productivity gains as the primary benefit. The pattern of adoption (starting with basic automation, analytics, and workflow tools before moving to more complex applications) mirrors the trajectory seen in more advanced markets.
The challenge is that large enterprises represent a small fraction of the total business population. The productivity story at national scale will be written by what happens in the mid-market and SME segments, not in the top 200.
Government investment is beginning to flow
Budget 2025 allocated $213 million for tuition, training subsidies, and STEM priority skills, a recognition from government that the skills gap is real and requires public investment to address. In February 2025, the government introduced a comprehensive Public Service AI Framework establishing a roadmap for safe AI deployment across agencies. The national AI strategy, while late by international standards, does provide a coherent framework for investment, regulation, and capability building that was previously absent.
These interventions will take time to flow through to business outcomes, but they signal a shift in policy seriousness that is important context for business planning.
Early adopters are demonstrating the returns
Across New Zealand, organisations that have moved early are generating measurable results that are beginning to change the conversation. In agriculture, AI-powered monitoring and predictive analytics are improving yield and reducing input costs on farms. In professional services, firms using AI for document review, research, and client reporting are completing work faster and at lower cost. In retail and logistics, automation of inventory management and demand forecasting is reducing waste and improving service levels.
The evidence from early adopters is making the ROI case for the next wave of adopters, and the next wave is beginning to move.
Curious what AI ROI looks like in your industry?
We can walk you through real examples from comparable NZ businesses and help you build a realistic business case.
What New Zealand businesses must do right now
Catching up does not require a massive technology programme or a large upfront investment. It requires structured thinking, deliberate decisions, and the willingness to start. Here is a practical framework for any New Zealand business ready to close its AI gap.
Start with a use-case audit
The most common mistake businesses make is starting with the technology instead of the problem. Before evaluating any AI platform or tool, map the processes in your business where time is lost, errors occur, or bottlenecks slow decisions. Identify the five to ten workflows that are high-volume, rule-based, or data-heavy. These are almost always the most tractable AI opportunities.
A good use-case audit does not take months. A structured two-week discovery with the right facilitation can surface the top opportunities, rank them by effort and impact, and give you a prioritised roadmap for the next 12 months. That roadmap is the foundation of everything else.
Close the skills gap urgently
Skills investment does not have to mean hiring, though eventually it should. Start by ensuring your leadership team has enough AI literacy to make sound strategic decisions: what is genuinely possible, what is hype, what questions to ask vendors. Then build capability in the middle of the organisation, ensuring that team leaders understand how AI tools integrate with their workflows and how to evaluate quality.
Front-line training should be practical and role-specific, not generic AI awareness sessions. Workers need to know how to use the specific tools relevant to their jobs, how to verify outputs, and how to flag issues. Budget 2025's training subsidies make this more accessible than it has been. Use them.
Build governance before you scale
The 50%+ of organisations experiencing shadow AI are building risk exposure that will need to be unwound later, at greater cost and disruption than addressing it now. Establish a clear AI policy that covers which tools are approved, how data is handled, what outputs require human verification, and how the organisation will monitor for bias or errors.
Governance does not need to be bureaucratic. A concise, well-communicated AI usage policy that staff actually understand is far more effective than a lengthy document no one reads. The goal is to create a culture where AI use is visible, structured, and improvable, not invisible and unmanaged.
| Action | Timeframe | Who leads it | Expected outcome |
|---|---|---|---|
| Use-case audit | 2–4 weeks | Leadership + ops | Prioritised AI roadmap |
| AI literacy for leadership | 1–2 days | External facilitator | Confident strategic decisions |
| Internal AI policy | 2–3 weeks | IT + HR + Legal | Shadow AI reduced; risk managed |
| Pilot project (1 use case) | 4–8 weeks | Operations + IT | Proven ROI; team confidence built |
| Scale and expand | 3–12 months | Cross-functional | Sustained productivity gains |
Choose the right implementation partner
Most New Zealand businesses do not have the in-house expertise to run a successful AI implementation from scratch, and they do not need to. What they need is a partner who understands the local context, has a track record of delivering in comparable businesses, and is willing to build capability within the client organisation rather than create dependency on external support.
When evaluating partners, look for evidence of domain expertise in your sector, a structured methodology that starts with business problems rather than technology, and a transparent approach to data governance. The best implementations leave your team more capable at the end than at the beginning. Be cautious of any partner who leads with technology features rather than business outcomes.
Ready to turn AI strategy into action?
DataForge works with New Zealand businesses to design, pilot, and scale AI implementation, starting from where you are, not where you wish you were.
Frequently asked questions
Is New Zealand really behind in AI adoption?
Yes, by most credible measures. New Zealand was the final OECD country to publish a national AI strategy, 68% of NZ SMEs have no AI plans versus 38% in Australia, and New Zealand's overall AI adoption rate of 37.6% lags significantly behind Singapore (70%+) and most comparable advanced economies. Half of NZ business leaders themselves believe New Zealand lags other countries on AI innovation.
What is the economic cost of slow AI adoption in New Zealand?
Microsoft's 2025 AI Economy Report identified a NZ$3.4 billion near-term opportunity. The government's national AI strategy projects AI could add NZ$76 billion to the economy by 2038, but only if adoption accelerates substantially. Every year of delay is compound productivity and competitiveness lost.
What are the main barriers to AI adoption for NZ businesses?
The three most consistently cited barriers are: lack of internal skills and capability (32%), data security and compliance concerns (57%), and financial constraints (34%). Underpinning all of them is a governance vacuum: only 55% of organisations have clear AI policies, yet over half are already experiencing shadow AI use by staff.
Is it too late for New Zealand businesses to catch up?
No. AI capabilities are still early in their deployment cycle, and the structural advantages of early enterprise adoption are not yet irreversible. New Zealand businesses that move deliberately over the next 12–18 months can close much of the current gap. The cost of waiting, however, is rising every quarter as competitors (both domestic and international) build deeper capability and process maturity.