The UK government has admitted that AI adoption among businesses is below where it expected. Only around a third of small and medium-sized businesses are using AI tools, and nearly three quarters of those not using it say they have not even identified a need for it. Something is clearly not landing. But the problem is not the technology. It is the fog around it.

The Numbers Tell a Strange Story

Between 2024 and 2026, AI adoption among UK small and medium-sized businesses climbed from roughly 25% to somewhere between 35% and 39%. That sounds like progress until you look at the barriers. According to government research, 60% of businesses cite limited AI skills and knowledge. 72% point to regulatory uncertainty. And the single biggest reason businesses give for not adopting? They have not identified a need for it.

That last one is the telling figure. It is not that business owners looked at AI and decided it was not useful. It is that they looked at the noise around AI and could not see where it fit into their Tuesday morning.

Meanwhile, almost a third of small businesses say they do not know which digital tools they should be using. Not which AI tools. Which digital tools, full stop. The conversation has jumped ahead of where most businesses actually are.

The Regulation Question Nobody Can Answer Yet

On 18 March 2026, the UK government is required to publish an economic impact assessment on copyright and AI. This comes out of the Data (Use and Access) Act and covers how copyright law interacts with AI training data, what transparency requirements might look like, and how licensing arrangements could work going forward.

The report will address the full range of issues raised during a public consultation that ran from December 2024 to February 2025. That includes how copyright affects access to training data, technical standards, enforcement, and questions about AI-generated outputs, including labelling and digital replicas.

For a business owner, here is what that means in plain English: the rules around AI-generated content are still being written. Nobody has done anything wrong by using these tools. But the government is actively deciding what the guardrails will look like, and until that picture is clearer, a lot of businesses are understandably cautious.

The honest position is this. Everyone wants to respect copyright. Everyone also wants to use the tools available to them. Those two things are not in conflict right now, but it is worth being aware that there is an active discussion about where the lines sit. If you are using AI to generate content, proposals, or marketing materials, just be mindful of that when you review the outputs. Do not copy and paste blindly. Apply judgement. That is good practice regardless of what the regulation ends up saying.

The Misconception That Is Costing Businesses Time

Here is where the real problem sits. Most of the conversation right now treats “AI” and “automation” as the same thing. They are not.

AI is a specific subset of automation. It involves language models, pattern recognition, and decision-making based on probability. It is powerful. It is also more complex, more expensive to run, and harder to maintain than straightforward process automation.

The majority of things a business with 10 to 50 employees wants to automate do not need AI at all. Moving data from one system to another. Sending a follow-up email three days after a quote is issued. Generating a weekly report from numbers that already exist in a spreadsheet. Flagging an overdue invoice. Routing a form submission to the right person.

None of that requires a language model. It requires logic: if this, then that. It is deterministic. It is predictable. And because it does not involve AI, it is cheaper to build, more reliable to run, and not touched by any of the regulatory discussions happening right now.

The irony is that AI opened the door to the automation conversation. Business owners who had never thought about automating their processes started paying attention because AI was suddenly in every headline. That is genuinely useful. But somewhere along the way, the message became “you need AI” when for most businesses, the message should be “you need to stop doing things manually.”

Where AI Actually Fits

None of this is to say AI is irrelevant to small businesses. It is not. But it sits at the more advanced end of the automation spectrum, and it helps to be specific about where it adds genuine value.

AI is useful when the task involves unstructured information. A customer email that could mean five different things. A document that needs summarising. A support query that needs triaging based on tone and content, not just keywords. Draft responses, content generation, data extraction from messy sources. These are real, practical applications where a language model does something that rule-based automation cannot.

The point is sequence. Most businesses would get more immediate value from automating the straightforward, repetitive work first. That clears time and headspace. It also builds confidence in automation as a concept, which makes the step up to AI-assisted processes far less daunting when the time comes.

And by then, the regulatory picture will be clearer too.

What the Next Few Months Will Tell Us

The 18 March report is not going to resolve every question. It is an economic impact assessment, not final legislation. But it will signal direction. It will tell us whether the government is leaning toward stricter licensing requirements for AI training data, lighter-touch transparency obligations, or something in between.

For businesses waiting on the sidelines, that signal matters. Not because it changes what you should automate today (it does not), but because it will start to clarify the rules for the AI-specific applications you might want to explore six or twelve months from now.

In the meantime, the 72% of businesses citing regulatory uncertainty as a barrier might want to ask a different question. Instead of “what are the rules around AI?”, try “what am I doing manually that does not need AI to fix?”

The answer to that question is not waiting on any government report. It is sitting in your inbox, your spreadsheets, and the processes your team runs on muscle memory every single week.

The Gap Is Not Regulatory. It Is Knowledge.

The government’s own research makes this clear. The biggest barrier is not regulation. It is that businesses do not know where to start. They hear “AI” and picture something complicated, expensive, and legally uncertain. So they do nothing.

The reality is simpler. Most of the automation that would make a genuine difference to a 20-person business is rule-based, affordable, and completely unaffected by the copyright debate. AI adds value on top of that foundation, for specific tasks, when you are ready for it.

The businesses that will pull ahead in 2026 are not necessarily the ones adopting the most sophisticated AI. They are the ones that stopped doing things by hand.

If you want to work out where your business is losing time to manual processes, drop me an email at [email protected]. No pitch, just a practical conversation about what could be automated.