Friday, April 10, 2026 3:02 p.m.
British businesses are rapidly adopting AI, but many are still struggling to turn that momentum into real and meaningful economic benefits.
According to IBM’s Race for ROI study, 66% of UK businesses report significant productivity gains from AI, with 63% of senior leaders citing significant efficiency gains since implementation.
There appears to be a gap between early successes and broader transformation. However, as the same study shows, 62% of organizations are still unable to realize the full potential of AI. And that disconnect extends across boardrooms and business units alike.
“This is not a technology issue; it’s an organizational transformation issue,” says Sebastian Weir, executive partner at IBM.d City am. “You can’t give your employees access to a co-pilot and expect them to just go along with it.”
AI implementation and scale comparison
More than three-quarters of UK businesses use some form of AI tools, and IBM research shows that businesses are already benefiting from time savings and increased efficiency.
These advances are translating into tangible changes in the way we work, with IBM reporting that AI frees up time for more high-value tasks, such as innovation (41 percent) and creative work (41 percent).
But a study by consultancy Studio Graphene found that progress beyond early signs of gains has been uneven.
The report shows that while 31% of UK companies using AI achieve some positive return on investment, fewer than half can clearly define what success means.
Similarly, a survey of global companies suggests that fewer than one in 10 companies are seeing a significant financial impact from AI.
“The friction we’re seeing is how do you begin to change behavior, from leadership KPIs to changing organizational behavior,” Weir says.
Without these changes, AI is likely to remain limited to isolated use cases rather than driving broader business transformation.
unrealistic expectations
A key barrier is the level of expectation that AI will deliver immediate economic benefits.
IBM research found that while more than a quarter of UK businesses have already realized cost savings, a further 34% expect to see benefits within the next year, as the effects often take time to show.
“It’s unrealistic to expect big profits right away,” Weir argues. “It’s not that the benefits don’t exist, it’s just that we are realistic about the future.”
Much of the initial value of AI is indirect, with efficiency gains such as reduced time spent on daily tasks difficult to measure in traditional quantifiable monetary terms.
“Even if you save three minutes on the phone, it’s very difficult to convert that savings into cash,” he added. “But it makes me more efficient.”
This creates tension at the board level, where investment decisions are often tied to short-term performance metrics rather than long-term productivity gains.
Skills and model bottlenecks
In addition to measurement challenges, employee readiness remains a major constraint.
An IBM study found that 67 per cent of UK business leaders cite internal resistance and cultural barriers as key barriers to AI adoption.
And at the same time, only 38% of organizations are actually prioritizing improving AI skills across their workforce.
Furthermore, YouGov research commissioned by The Access Group found that this gap is reflected in the way employees use technology. 70% of UK employees have tried AI tools, but only 19% have received formal training.
“There were quite a number of organizations that released AI guidance three years ago that said, ‘Don’t do this,'” Weir added.
“Everyone needs critical thinking skills to understand what is right, what is wrong, and what to trust.”
New systems and updates are released frequently, making it difficult for businesses to keep up with the speed at which AI models are advancing.
While this encouraged investment, it also made it difficult for organizations to stay focused on results.
“It’s incredibly difficult to keep up. Frontier models are evolving faster than anyone can actually consume them,” Weir said.
If teams prioritize technology upgrades over delivering results, the pace of change can stall projects.
“It’s very easy to see how a program stagnates when you’re focused on models rather than results. Many of our projects are built on AI models that are three to four years old, but they’re still great and can make real, tangible changes,” he added.
IBM’s broader Enterprise 2030 study found that while 79 percent of executives expect AI to make a significant contribution to the bottom line, only 24 percent can clearly identify where that growth will come from.
A worrying gap between expectations and execution could define the next stage of AI adoption.
Weir points to alignment of governance, measurement and leadership as key areas for improvement. “If you haven’t changed your KPIs, your behavior hasn’t changed,” he said.
