New Zealand firms split on AI as strategy gap widens

Amazon Web Services has published research showing a widening gap in AI adoption among New Zealand businesses. The survey found that half now use AI consistently.

While AI use is becoming more common, only 15% of businesses have a formal AI strategy. Those with a strategy were more than twice as likely to have reached advanced adoption: 41%, compared with 19% of AI adopters overall.

The study, titled Unlocking New Zealand's AI Potential 2026, surveyed 1,000 businesses and 1,000 individuals across the country. It suggests a two-tier pattern is emerging, with a smaller group of organisations linking AI use to revenue, product development and expansion, while many others remain at an earlier stage.

Among businesses already using AI, 84% said it had increased revenue and 69% reported productivity gains. Yet only 34% were confident they could measure the return on investment from AI spending, and 61% had no dedicated AI budget.

Readiness for newer forms of AI also remained limited. Just 17% of businesses said they felt prepared for the next generation, including agentic AI, physical AI and advanced robotics.

Strategy gap

Formal planning was one of the clearest markers separating more advanced users from the wider market. Businesses with an AI strategy were nearly twice as likely to say they could measure return on investment, 63% versus 34%, and more than twice as likely to say they were ready for the next phase of AI adoption, 44% versus 17%.

The divide was not simply a matter of size. Startups were ahead of both large enterprises and small and medium-sized businesses in advanced AI use, suggesting that speed of decision-making and clarity of purpose may matter more than scale alone.

The report also pointed to signs that AI is moving beyond back-office experimentation. Nearly a third of respondents, 32%, said they had launched a new AI-driven product or service in the past year, while 27% said AI had helped them expand internationally.

Examples in market

AWS highlighted several New Zealand businesses using AI in operational settings. Halter built an AI agent called Clank on Amazon Bedrock that, according to AWS, saved more than 215 hours of engineering time by helping to triage system alerts.

ZURU is using AI to analyse 20,000 social videos a day, helping cut product development time from 18 months to five, according to AWS. Parrot Analytics is using Amazon Bedrock to verify content data 10 times faster at 60% lower cost, while Tower Insurance, working with Deloitte, deployed an AI-based contact centre in 10 weeks and saved more than one million minutes of customer call time, AWS said.

Manuel Bohnet, Country Manager, AWS New Zealand, said customer conversations had shifted over the past year. "The conversations I'm having have shifted from 'how do I use AI?' to 'how do I use AI to change my business?'" Bohnet said.

Skills pressure

The report identified workforce capability as a major constraint. A lack of skilled personnel was the leading barrier to AI adoption, with digitally skilled roles taking an average of 3.4 months to fill.

Demand for broader AI literacy is also rising. The research found that 52% of new roles are expected to require AI literacy within three years, while 54% of New Zealanders said they had used an AI tool to teach themselves a new skill in the past year.

By contrast, only 34% said they had received AI or digital training from their employer. That mismatch suggests individual learning is moving faster than workplace investment in training.

AWS said it has trained more than 100,000 New Zealanders in cloud skills through programmes including AWS Skills to Jobs Tech Alliance, AWS Academy, AWS Educate and the AWS Generative AI Accelerator. The company is also introducing a new certification category for business decision-makers, beginning with AI Business Strategist.

Economic divide

The broader message from the research is that AI uptake alone is no longer the most revealing measure of progress. With half of businesses now reporting AI use, the more important distinction appears to be whether organisations can identify where AI creates value, assign budget, measure outcomes and scale successful projects.

Bohnet said New Zealand's challenge was no longer ambition. "The challenge is now turning that ambition into capability across the economy," he said.