In short: AI use among UK businesses has almost tripled since late 2023, and around one in three now use it. But the latest ONS figures hide a bigger number: 55% of employees say they use AI for work, 20 points ahead of their employers. Staff have adopted AI on their own, while their employers have mostly not bought, trained for, or integrated anything. The businesses that close that gap will be the ones that stop treating AI as a gadget and build it into how they run.
For three years, the story of AI in British business has been one of steady growth. Every quarter, more firms reported using it, and the line on the chart pointed firmly up. It still does.
According to the Office for National Statistics, 34.9% of UK businesses with 10 or more employees now use at least one AI technology. That is up from 11.9% in September 2023, an almost threefold rise. But the same release contains a number that changes how you read the first one.
Ask employees rather than employers and adoption looks very different. Over half of workers (55%) say they use AI for work or education. In other words, AI has arrived in most UK workplaces, but it came in through the staff rather than the business. You can see the full picture in our study of UK AI adoption statistics.
The Boom Is Real
The AI rise has been steep. In late 2023, around 12% of UK businesses with 10 or more employees reported using AI. By June 2026 that had risen to almost 35%, and it has grown every quarter along the way. It’s no wonder we’ve seen headlines about an AI revolution sweeping through the economy.
The ONS figures put the last four readings at 27.2% in September 2025, 28.7% in December, 32.1% in March 2026, and 34.9% in June 2026. That is a rise of roughly three points a quarter, with no sign of slowing. A plateau, if one comes, is not in the data yet.
UK AI Adoption Has Risen Every Quarter Since 2023
| Survey wave | Businesses using at least one AI technology | Change on previous reading |
|---|---|---|
| September 2023 | 11.9% | First reading |
| June 2024 | 14.8% | +1.0 pt |
| June 2025 | 25.1% | +4.5 pts |
| September 2025 | 27.2% | +2.1 pts |
| December 2025 | 28.7% | +1.5 pts |
| March 2026 | 32.1% | +3.4 pts |
| June 2026 | 34.9% | +2.8 pts |
Businesses with 10 or more employees. Changes are shown against the previous quarterly reading in the ONS series. Source: Office for National Statistics, Artificial intelligence in UK businesses: 2023 to 2026 (Business Insights and Conditions Survey, Wave 159). Contains public sector information licensed under the Open Government Licence v3.0.
So what is the problem? The headline counts a business as an adopter if it uses a single AI technology anywhere in the organisation. On that measure, adoption is rising fast. On almost every other measure, the picture is thinner.
Staff Have Adopted AI. Their Employers Have Not
Here is the number that matters. The ONS compares its business survey with its household survey, and the two disagree by 20 points. Some 55% of employed and self-employed people say they use AI for work or education. Only 35% of businesses say they use it.
The gap is not a measurement error. It is a description of how AI is arriving in most UK workplaces: through individuals, using free tools, without a purchase order. When manufacturers do adopt, free-to-use software is the single most common route in, used by 49% of them.
How AI-Using Businesses Got Their AI, by Sector
| Sector | Free-to-use software | Purchased software or ready-to-use | Developed in-house | Outsourced to a provider |
|---|---|---|---|---|
| Manufacturing | 49.1% | 34.6% | 14.9% | 7.9% |
| Wholesale and retail | 53.6% | 41.3% | 5.9% | 3.4% |
| Construction | 26.1% | 41.8% | 4.4% | 8.2% |
| Accommodation and food | 54.3% | 12.5% | 2.5% | 2.9% |
| Admin and support services | 48.3% | 37.2% | 6.0% | 13.7% |
| Professional and scientific | 40.6% | 49.6% | 14.5% | 11.3% |
| Information and communication | 50.8% | 56.7% | 28.5% | 4.9% |
| All businesses | 44.5% | 42.8% | 11.6% | 7.9% |
Percentage of businesses that report using at least one AI technology, or are not sure whether they do. Businesses can select more than one method, so rows do not sum to 100. Source: Office for National Statistics, Business Insights and Conditions Survey, Wave 159 (15 to 28 June 2026). Contains public sector information licensed under the Open Government Licence v3.0.
That has consequences. Among businesses that do use AI, 46% have given no staff any training in it, and only 16% use it extensively. Two in five businesses with 10 or more employees say nothing has prevented or delayed their adoption. They are not blocked, they simply have not decided.
Why Businesses Are Slower Than Their Staff
The first wave of adoption was the low-hanging fruit. Think individuals and teams trying a chatbot or a text generator, often off their own back. It’s quick and cheap, requiring no real change to how a business runs. The next stage means embedding AI into actual operations, and that is where most businesses stall.
The barriers aren’t really about the technology either. In the ONS survey, a lack of expertise is the most cited barrier in every size band, ahead of cost. Among businesses with 100 to 249 employees, 18% say limited AI skills have delayed adoption, against 14% citing cost. Across 2026, survey after survey has found that the things holding businesses back are:
- A lack of skills
- A lack of clear use cases
- Doubts about whether the spend is worth it
What Is Holding Businesses Back, by Business Size
| Business size | No barriers reported | Lack of AI expertise | Cost | No clear use case | Lack of trust in AI | Security concerns |
|---|---|---|---|---|---|---|
| 0 to 9 employees | 51.6% | 7.0% | 6.5% | 5.4% | 6.4% | 4.0% |
| 10 to 49 employees | 42.0% | 13.3% | 11.3% | 8.7% | 8.0% | 6.2% |
| 50 to 99 employees | 35.3% | 17.0% | 12.8% | 11.3% | 10.0% | 10.4% |
| 100 to 249 employees | 32.4% | 18.2% | 13.7% | 10.9% | 10.1% | 9.7% |
| 250 or more employees | 33.6% | 14.6% | 11.0% | 8.8% | 7.9% | 9.1% |
Percentage of businesses reporting each factor as having prevented or delayed AI adoption in the last three months. Businesses can report more than one barrier. “Other”, “not sure” and “has not attempted to use AI” are excluded. Source: Office for National Statistics, Artificial intelligence in UK businesses: 2023 to 2026, Figure 10 (Business Insights and Conditions Survey, Wave 159). Contains public sector information licensed under the Open Government Licence v3.0.
Government-commissioned research points the same way. The Department for Science, Innovation and Technology found that 71% of businesses had not identified a use for AI and 60% cited limited AI skills. In short, AI comes across as another standalone tool to learn, justify, and secure. With that in mind, it’s easy to see why so many firms simply stop.
There’s also a confidence problem. A lot of what businesses read about AI is noise: agentic this, autonomous that, a constant sense of being five years behind. When they’re faced with that, plenty of firms freeze rather than act.
It doesn’t help that a lot of the early adoption has been shallow, with the odd subscription or licence but little to show for it. We’ve written before about the common myths that put businesses off AI. Most of them are exactly the kind of thing that keeps a business watching from the sidelines.
The Average Hides a Widening Gap
A national average can hide a lot of movement underneath. In this case, it’s hiding a gap that’s getting wider. The largest firms have kept adopting: 49% of businesses with 250 or more staff now use AI. Among the smallest, those with fewer than 10 employees, it is 28%.
The sector picture is just as uneven. Information and communication businesses sit at 58% adoption, and professional services at 46%. Manufacturing is at 22% and construction at 13%.
Who Has Adopted AI, and How Far They Have Come Since 2023
| Business size | Using AI, September 2023 | Using AI, June 2026 | Change |
|---|
Percentage of businesses reporting use of at least one AI technology. Size and sector rows cover all business sizes within that group. The total row is the ONS headline series for businesses with 10 or more employees. Sector samples are smaller, so figures move more between waves. Source: Office for National Statistics, Business Insights and Conditions Survey, Wave 159 (15 to 28 June 2026). Contains public sector information licensed under the Open Government Licence v3.0.
Analysis of the same ONS data has described UK adoption as widening rather than deepening. The average adopter uses barely more AI tools than it did in 2023: 1.6 against 1.4.
So, the picture is really two stories:
- Digitally mature firms and their staff pulling ahead
- Everyone else watching, while their own employees quietly use AI anyway
For the businesses in that second group, the risk is that this year’s gap becomes next year’s chasm.
How to Close the Gap
If the first wave of adoption came from staff bolting free tools onto their own workflows, the next phase has to come from something sturdier. We’re talking about AI built into the systems a business already runs on.
That’s the difference between an AI experiment that fizzles out and one that sticks. When AI lives inside your finance, sales, and stock system, there’s no separate tool to learn and no shadow usage to police. Crucially, there’s a much clearer line between the spend and the return.
This is exactly the territory where Copilot in Business Central changes the maths. Because the AI is part of the platform your team already uses, adoption is a setting rather than a leap. It matches invoices, answers plain-English questions about your numbers, and clears routine work off people’s desks.
If your staff already use AI but your business does not, moving beyond AI experiments and into real operations closes that gap. It also turns a governance risk into an advantage.
Some people read the adoption numbers and see a country that has already embraced AI. What they actually show is a workforce that has, and a set of businesses that have not caught up. The firms that treat this as the moment to go deeper will pull ahead while everyone else waits.
Catch Up with Your Own Staff
If your team is already using AI and your systems are not, the fix is a platform that puts AI to work where it counts. Let us show you how Business Central builds AI into the way you already run finance, stock, and operations. Speak to the team about turning scattered personal tools into real momentum.
Frequently Asked Questions
How Many UK Businesses Use AI in 2026?
The ONS puts it at around one in three businesses with 10 or more employees: 34.9% as of June 2026. Across all business sizes the figure is 29%. Surveys quote different figures because they measure different business populations, but the ONS is the most consistent national measure over time.
Is AI Adoption in the UK Still Growing?
Yes. ONS figures show adoption rising every quarter since late 2023, from 11.9% to 34.9% of businesses with 10 or more employees. The last four readings were 27.2%, 28.7%, 32.1%, and 34.9%, roughly three points a quarter. What has barely moved is depth: the average adopter uses 1.6 AI technologies, up from 1.4.
Why Do More Employees Use AI Than Businesses?
Because staff can pick up free tools without anyone signing anything off. The ONS finds 55% of employees use AI for work or education, against 35% of businesses reporting adoption. Much of that use sits outside anything the business has bought, trained for, or governs.
Which Businesses Are Adopting AI Fastest?
Mostly the largest and the most digital. Around 49% of firms with 250 or more staff use AI, compared with 28% of the smallest. Information and communication sits at 58%, while manufacturing is at 22% and construction at 13%.
How Can a Business Catch Up with Its Staff?
By building AI into the systems it already runs rather than adding more separate tools. AI inside a core platform like Business Central with Copilot removes the learning curve and makes the return far easier to see. It also brings shadow usage under control.
Where Does This Data Come From?
The business figures are from the ONS Business Insights and Conditions Survey, Wave 159 (June 2026). They are supported by the ONS article “Artificial intelligence in UK businesses: 2023 to 2026”. The employee figure is from the ONS Opinions and Lifestyle Survey. You can read our full breakdown in our UK AI adoption statistics study.
