Poor progress monitoring and poor follow-through on initial training were factors. Poor training also came through in a PSA survey earlier this week.

“We’re hearing a lot of urgency from boards and leadership. But do middle managers and employees have a change plan in place?” McFadden asked.

In terms of AI readiness, New Zealand has slipped back, according to the Cisco survey.

The survey found 19% of NZ firms were “pacesetters” or fully prepared for AI, while 13% were “chasers” or moderately prepared (the balance were “followers” with limited preparedness or “laggards” who were unprepared).

The percentage of Kiwi firms in the leading categories fell from 38% to 32% year-on-year (the overall global survey had 46% of firms fully or moderately prepared).

Organisations are spending heavily on AI, with 42% of companies allocating 10-30% of their current IT budget to AI deployment, according to the survey.

But despite significant AI investments in strategic areas like cyber-security, IT infrastructure, and data analytics and management, many companies report that returns on these investments are not meeting their expectations, McFadden said.

Nevertheless, the pressure is on for results.

Nearly all (97%) of respondents said they were under more chief executive or boardroom pressure to produce results from AI projects.

Of those surveyed, 58% thought they had a maximum of one year to implement their AI strategy before significant negative business impacts; the number rose to 83% with an 18-month timeframe.

But for many, it was not clear how they would develop that strategy, with budget, talent shortages, cybersecurity, quality of in-house data, and infrastructure deficiencies identified as pain points.

Only 29% of the respondents in New Zealand said AI deployment has been given the highest priority in their budget allocation and incremental budget funding, compared to the other technology deployments that they’re working on.

The result is in tune with an early survey by the Technology Users Association of New Zealand, which found many IT leaders feeling the heat to introduce AI – but within budgets squeezed by the economic slowdown, with money for new projects often reduced.

“As the race to adopt AI picks up pace, talent will be a key differentiator for companies,” McFadden said.

“There is already a shortage of skilled talent across various aspects of AI. This means companies will need to invest in their existing talent pool to meet the growing demand.

“At the same time, it is crucial that all stakeholders – the private and public sectors, educational institutions, and governments – work together to develop local talent so that the entire ecosystem can benefit from the immense potential that AI offers.”

The current Government has so far not allocated any new funding toward public or private AI initiatives, although some existing funding has been reallocated with Callaghan Innovation’s AI Activator programme, which aggregates existing Crown resources, making them easier to find.

Callaghan is being defunded after June 30. Legislation enabling a new public research organisation, which will take on some of Callaghan’s role, will be passed in the final quarter of this year and come into effect in early 2026, the Government said.

Chris Keall is an Auckland-based member of the Herald’s business team. He joined the Herald in 2018 and is the technology editor and a senior business writer.