New Zealand firms struggle to scale AI across business

Most New Zealand organisations are struggling to scale artificial intelligence across their operations, according to Datacom's latest AI Index. Just 4% say AI has transformed core operations.

That is down from 8% a year earlier, even though 91% of organisations now use some form of AI.

The survey highlights a widening gap between adoption and deeper operational change. AI use has risen steadily from 66% in 2024 to 87% in 2025 and 91% now, yet 81% of organisations remain in the exploratory or implementation stages of AI maturity.

Only 15% say they are scaling AI across the organisation, up modestly from 12% a year earlier. The sharper shift came at the top end of maturity, where the share of businesses saying AI had transformed core operations halved.

Peter Nelson, Managing Director, Datacom New Zealand, said the findings show many companies still treat AI as a limited workplace tool rather than a broader business issue.

"The data suggests organisations are finding it much easier to introduce AI for basic everyday use cases than to scale it across their business. AI adoption has accelerated dramatically over the past two years, but widespread use isn't automatically translating into organisation-wide change.

"What we're seeing in our work with customers is that the organisations making the most progress are treating AI as a business priority, with clear ownership, clear objectives and a strong focus on where AI can create value."

Leadership gap

One of the clearest weaknesses in the research is leadership structure. Just 22% of businesses have a dedicated AI leadership role, including 13% with a Chief AI Officer and 9% with another full-time AI leadership position.

In many organisations, responsibility sits elsewhere. Some 26% say AI has been added to an existing executive role such as Chief Information Officer, Chief Technology Officer or Chief Digital Officer, while 19% say responsibility is shared across several executives.

Lou Compagnone, Director of AI, Datacom, said that arrangement could slow progress once businesses move beyond early trials.

"For many organisations, AI is still treated as a technology initiative, with responsibility added to an existing executive role. But AI now touches strategy, workforce capability, governance, risk and organisational change.

"If organisations want to move beyond pilots and isolated use cases, they need clear ownership and accountability. The organisations making the most progress are approaching AI as a business transformation priority rather than a technology project."

Skills remain another constraint. When asked to identify the main barrier to scaling AI, 20% cited a lack of internal capability or skills, making it the most common response.

Employee concern, resistance or fear of job displacement followed at 18%, with implementation costs close behind at 17%. The mix of concerns has shifted from a year earlier, with internal capability becoming less dominant while workforce anxiety and costs have become more prominent.

There are signs employers are trying to close the gap. Fifty-nine per cent now either employ or plan to employ dedicated AI talent, up 25 percentage points from the previous year, while 66% have provided AI skills training over the past 12 months, up 10 points.

Compagnone said the issue is increasingly about redesigning work rather than simply adopting software.

"The biggest gap we're seeing isn't between organisations using AI and those that aren't. It's between organisations that have adopted AI and those that are adapting their operating model around what AI now makes possible.

"Using AI tools is one thing, but knowing what to shift from and to - from existing workflows, services and ways of working to new models designed around AI - is something else entirely, and that's where the real value will come from."

Investment trends

Spending continues to rise despite uneven progress. Datacom found 79% of New Zealand businesses increased AI investment over the past year, and 73% expect to invest more over the next 12 months.

Some companies are also reporting financial benefits. A third say returns from AI have exceeded costs, while another 30% say benefits and costs are broadly balanced.

Time savings and productivity gains were the most commonly cited source of value at 59%. Improved quality and accuracy of outputs came next at 41%, followed by operational efficiency at 34%.

Use remains concentrated in mainstream tools. General-purpose AI assistants, including ChatGPT, Gemini and Claude, were the most commonly used technologies at 68%, followed by AI-enhanced productivity tools at 66%.

More specialised forms of deployment were less common. Some 31% reported using AI embedded in business platforms, 26% used custom-built or in-house systems, and 13% used agentic or autonomous AI systems.

Strategy and governance appear to be developing, but unevenly. While 62% of businesses say they have some form of AI strategy, only 16% have a clearly defined standalone strategy. Another 23% have formally embedded AI within a wider business, digital or technology strategy, while 23% say their strategy remains high-level or is not yet fully operationalised.

Cost discipline is also still taking shape. One in five organisations say they have not yet addressed AI cost management, and only 20% have a dedicated AI budget with clear cost tracking.

Policy pressure

The findings also suggest growing demand for a clearer public policy framework. More than half of business leaders, 55%, say they are somewhat or very confident that New Zealand is prepared to manage the risks and opportunities of AI, up 18 percentage points from a year earlier.

At the same time, support for more formal oversight is strong. Nearly two-thirds, 64%, say New Zealand should introduce specific AI legislation and controls, while 74% back a dedicated national AI framework and coordinating body.

Nelson said businesses want clearer rules as AI moves closer to critical functions.

"Businesses are looking for greater clarity around the environment they will be operating in. As AI becomes more embedded in critical services and everyday operations, organisations want confidence that there are clear expectations around governance, accountability and risk.

"That certainty helps organisations innovate more confidently and invest for the long term."

Compagnone said the next stage will depend on ownership and execution rather than basic uptake.

"The organisations creating the most value from AI are identifying where it can have the greatest impact, putting clear ownership around those opportunities and building the capability to scale them across the business."