AI is everywhere in the headlines, from HSBC’s reported plans to reshape its UK wealth business around it to a new wave of AI-discovered cyber threats. Yet the latest official figures on AI adoption show most UK businesses are still not using it, and many that are have barely scratched the surface. Here is what the data really says and how to turn AI into measurable results.

What the latest data shows

The Office for National Statistics‘ latest Business Insights and Conditions Survey, published on 8 October 2026, found that 30% of UK businesses were using at least one AI technology in September 2026. That level of AI adoption is up from 9% when the question was first asked in September 2023, but broadly flat since June.

Size makes a big difference to AI adoption. Among businesses with 10 or more employees, adoption reached 38%. Around half of the largest firms now use AI, compared with just 29% of businesses with fewer than 10 staff. By sector, AI adoption is highest in the information and communication industry, where around 60% of businesses use AI.

What businesses are using is telling. Among firms with 10 or more employees, the most common tools were text generation using large language models, at 21%, and visual content creation, at 19%. In other words, much of today’s adoption is still about chat assistants and content tools rather than AI built into core operations.

Why do AI adoption surveys give different numbers?

If you have seen higher figures elsewhere, you are not imagining it. The British Chambers of Commerce, working with Atos, reported in March 2026 that 54% of the SMEs it surveyed were actively adopting AI, up from 35% a year earlier. The difference comes down to who is asked and how the question is framed. The ONS survey is a large, weighted sample of UK businesses asked about specific AI technologies, while other surveys often capture firms that are exploring or experimenting. Both point the same way: AI adoption is rising fast, but unevenly.

The real problem: AI adoption without integration

The bigger gap is not whether businesses use AI but how deeply. The Department for Science, Innovation and Technology’s UK Business Data Survey found that only 21% of AI-using businesses said their AI tools were integrated into existing platforms such as Microsoft 365, CRM or finance systems, rising to 57% among large businesses. Governance is thin too: only 17% of AI-using businesses had any AI policy, and just 5% had a formal written one.

Security is lagging as well. The government’s Cyber Security Breaches Survey 2025/2026 found that only 24% of businesses using AI had any cyber security practices in place to manage the associated risks.

Is AI taking jobs?

For most UK businesses, not yet. The ONS found that just 6% of AI-using firms with 10 or more employees reported a lower headcount as a result, and fewer than 1% reported an increase. The BCC found that 95% of SMEs using AI had seen no impact on workforce size over the past year.

But the work itself is changing. Administrative and clerical roles were the most affected, cited by 38% of AI-using firms, up nine percentage points since December 2025. Creative and design roles came next at 24%. And at the large-company end, change can be sharp. The Financial Times reported this month that HSBC plans to cut around half of the management and specialist roles in its UK wealth business, and up to around 70% of its financial advisers, as it puts AI at the centre of how it serves clients. HSBC has said it is evolving to offer more digitally enabled products and journeys, and reports its third-quarter results on 27 October.

6 ways to move beyond the pilot

1. Start with a business problem, not a tool

Pick two or three processes where time, cost or errors genuinely hurt, such as invoice handling, customer queries or reporting. Define what success looks like in numbers before you choose any technology.

2. Get your data ready

AI is only as good as the information it can reach. Clean up and connect the data behind your chosen processes, and make sure access is controlled so the right people and systems see the right information.

3. Build AI into the systems people already use

The biggest returns come when AI works inside your existing workflows, in your email, CRM, finance or service desk tools, rather than in a separate window someone has to remember to open.

4. Write an AI policy, then keep it short

With only 17% of AI-using businesses having any policy, a clear one-page guide on approved tools, what data can be shared and who is accountable will already put you ahead of most of the market.

5. Secure it from day one

Assess the cyber risks of any AI tool before you deploy it, not afterwards. Control access, log usage and make sure sensitive data is not leaving your organisation through unapproved services.

6. Invest in people as well as platforms

The roles most affected by AI are changing, not simply disappearing. Train teams to use AI well, redesign roles around the new workflows and bring staff into the process early. AI adoption sticks when the people doing the work trust the tools.

The bottom line

Three in ten UK businesses have started on AI adoption, but far fewer have made it part of how they actually work. The gap between experimenting and transforming is where the real value sits. Firms that close it, with clear goals, connected data, sensible governance, and well-supported teams, will set the pace for everyone else.

Ready to move your AI from experiment to results?

Cloud Technology Hub helps UK and international organisations identify high-value AI use cases, prepare their data and integrate AI securely into the systems their teams already use.

Speak to our team at info@technohub.cloud. TALK

Leave a comment and Read More Here