More than eight-in-ten business leaders are confident that the UK can lead in AI, but only half as many UK employees agree on that front, according to new research from Microsoft.

A survey of 5,000 UK employees and 1,000 of their bosses as part of Microsoft’s AI Diffusion Report reveals a growing gap between how business leaders and their employees see the potential benefits of AI.

Of the business leaders surveyed, 84% reported feeling confident that the UK can become “a world-leading economy within five years,” but only 41% of UK employees agreed.

Mind the AI gap

What’s behind the gap? According to Microsoft, 35% of the surveyed workers believed AI had reduced their own job security, with 57% saying it hurts job prospects for younger people.

Despite those concerns, 71% also agreed AI had a positive impact on their work and 69% believe it had boosted the quality of their work.

While business leaders and workers may disagree on the job risks and utility of AI, about six-in-ten of both groups believe the government should do more to help the UK reap the benefits of the technology. Indeed, more than a third from each group are now calling for heightened regulation and safeguards.

“Government and industry must work together to respond to British business leaders’ calls for greater action on skills, build trust in the technology and ensure its benefits reach everyone,” said Darren Hardman, CEO of Microsoft UK & Ireland.

“If we remain divided on whether AI will have a positive impact, if we fail to build trust in the technology and if we cannot put the necessary safety ecosystem for models in place, then we will not create the society we want.

“The UK has a chance to lead, not just in AI adoption, but in AI safety. Get both right, and we can unlock growth while ensuring the benefits of AI are open to all.”

AI diffusion in the UK

Microsoft’s wider AI diffusion report shows the UK has leapt from 35.6% adoption to 43.7% as of the second quarter of 2026. That ranks the UK second in the G7 when it comes to adoption rates after France, and eighth globally, according to Microsoft’s figures.

The report noted that this shows a disconnect between the countries making frontier AI and applying it, with the US and China centered in most discussions about the technology.

“This overlooks a distinction between frontier AI and the more prosaic routine, task-level AI use through which most productivity gains are likely to be realised: tools embedded across routine business processes, public services, and everyday work,” the report notes.

“Frontier capability does not automatically produce broad diffusion, as the United States’ comparatively low adoption rate of 33.0% illustrates, ranking at 21st globally in Q2 of 2026.”

Development vs diffusion

Notably, the study found that the UK can’t match the US or China in frontier capital spending, adding that an “excessive focus on that contest” risks distracting from how the UK can gain economically from the technology.

As a nation, the UK “does not need to be a leader in model development to reap the benefits”, Microsoft said. The report added that countries that use these technologies will see the greatest long-term gains.

Regardless, Microsoft said the UK should still strive to meet the government’s stated goal of becoming the country with the fastest AI adoption in the G7.

“Our thriving technology sector is playing a fundamental role in the economic growth recognised by the IMF,” Hardman commented. “Business leaders and employees across the country are embracing AI to change not only what they do and how they get things done, but how quickly.”

The report from Microsoft comes amidst increased investment in AI-related support for British businesses. The company has committed to £23 billion in investment over the next three years to help drive adoption rates, including $15bn in cloud and AI infrastructure.

However, recent reports show the company has pulled out of a data center deal with Nscale. As ITPro reported on 19 October, the deal has fallen through due to lingering concerns over power supply for the Essex-based site.

The Loughton development was set to deliver 50 MW of AI capacity, scalable to 90 MW, and to initially house 23,040 NVIDIA GB300 GPUs. It was expected to create 250 permanent jobs.

FOLLOW US ON SOCIAL MEDIA