Most UK firms now use AI — the gap is between using it and getting value from it
Adoption has passed half of UK firms. But most use off-the-shelf tools, and the ones seeing real change are doing something harder. Here's what separates them.
Research from the British Chambers of Commerce in early 2026 found that more than half of UK firms (54%) now use AI in some form, up from 35% a year earlier and 25% the year before that. With about nine in ten of those firms being SMEs, this is squarely a small-business story now. The interesting part isn't the headline number. It's the distance between firms that have adopted AI and firms actually getting something out of it.
Most use the easy stuff, and that's fine
The bulk of that 54% are using widely available tools like ChatGPT or Microsoft Copilot for everyday work: drafting, summarising, answering questions. Only around one in ten have gone further and built AI into their own processes. There's nothing wrong with starting on the off-the-shelf tools, and most firms should. But it's worth being clear-eyed that typing questions into a chatbot is a long way from AI doing real work across your business.
Where the real change shows up
The firms reporting real operational change, including the few seeing changes to how they staff, are mostly the ones that have tailored AI to their specific work rather than relying on generic tools. That's harder, costs more upfront, and needs your data and processes in decent shape first. It also carries more risk, which is exactly why it needs proper guardrails. Bespoke AI isn't automatically the right call for everyone. The point is that bigger results require bigger groundwork.
The barriers are predictable, and fixable
Businesses keep citing the same blockers: a skills gap, too many disconnected AI subscriptions, uncertainty about return on investment, and no clear rules for how AI gets used. None of those is a technology problem. They are about focus and governance. Picking one or two high-value tasks, measuring whether AI actually helps, and writing a simple usage policy solves most of them.
The bit everyone skips
Before pushing AI deeper into the business, the unglamorous work is getting your data and permissions tidy. A tool like Copilot respects the file access your people already have, so messy permissions mean it can surface information to the wrong people. Sort that first and AI is a productivity gain. Skip it and a helpful tool becomes a data-exposure problem.
What this means for your business
If you're in the half already using AI, the next step isn't more tools, it's getting value from the ones you have. Pick a couple of routine, time-eating tasks, measure the payback, and tidy your data and permissions before going wider. We help South West businesses do exactly that, so AI saves real time without opening new risks.
#WEARECOBALT
Ready when you are.
Tell us what's slowing your business down. We'll tell you exactly how we'd fix it — plainly, with no obligation.