Do these 3 things before closing this tab:
1Repair Windows errors before they cause bigger problems2Scan for outdated or missing drivers - takes under a minute3Clear out junk files and repair common Windows errorsiTechGuides is reader-supported. When you buy through links on our site, we may earn an affiliate commission. As an Amazon Associate I earn from qualifying purchases. Learn more
UK companies are moving from AI experiments to adoption, but the shift is uneven and usually shallow. A 2025 government survey found 16% of businesses with at least five employees already using AI, while the Office for National Statistics (ONS) recorded use at about 35% among businesses with 10 or more employees by June 2026. The difference reflects separate samples and definitions, not a single contradictory trend.
The harder question is accountability. Financial firms commonly put senior executives in charge, yet most name three or more accountable people or bodies. Adoption is accelerating; ownership of risk, oversight and outcomes is less settled.
How many UK businesses use AI?
The clearest national measures show rapid growth, with substantially higher uptake among larger firms and technology-intensive sectors.
| Measure | Finding | Scope and date |
|---|---|---|
| Current use | 16% | UK businesses with at least five employees, UK Government survey, 2025 |
| Planning adoption | 5% | Businesses with at least five employees, UK Government survey, 2025 |
| Neither using nor planning AI | 80% | Businesses with at least five employees, UK Government survey, 2025 |
| Current use by size | 36% large; 23% mid-sized; 14% micro | UK Government survey, 2025 |
| Change over time | About 12% to about 35% | Businesses with 10 or more employees, late 2023 to June 2026, ONS |
| Late-December position | 25% using AI; 15% planning adoption within three months | ONS business survey, late December 2025 |
The government study involved 3,500 telephone interviews with private-sector businesses employing at least five people between 12 February and 2 May 2025, plus 100 qualitative interviews in March to May. Results were weighted by business size and sector. ONS business survey data use different samples and an employee threshold of 10, so the percentages should be read as directional evidence rather than combined into one precise time series.
#1 Best Overall
Which industries are adopting fastest?
Information and communication leads the ONS figures, with AI use reaching 58% of businesses in June 2026. Construction was at 13%. The gap is consistent with the way digital, data and software-heavy work can be integrated without the physical-process changes required in many construction operations.
Across the wider business population, the government survey identified marketing, administration and IT as the areas most likely to use or plan AI. Businesses that have adopted or are preparing to adopt it commonly place AI in future strategy documents, pairing immediate implementation with intentions to increase investment over time.
Is adoption transforming businesses or adding pilots?
So far, the evidence points to expansion in reach more than deep transformation. Among ONS businesses with 10 or more employees that use AI, the average number of AI technologies rose only from about 1.4 in late 2023 to about 1.6 by June 2026. Only 10% of AI-using businesses described their use as extensive.
Quick wins for a faster PC:
Repair Windows errors before they cause bigger problemsFix Now →Scan for outdated or missing drivers - takes under a minuteDriver Scan →“The self-reported use of artificial intelligence (AI) in UK businesses with 10 or more employees has increased from around 12% to around 35% since late 2023, with larger firms more likely to have adopted AI.”
Office for National Statistics, Artificial intelligence in UK businesses: 2023 to 2026, 2026
The same ONS analysis describes adoption as “relatively shallow”. A business may therefore be counted as an AI adopter because it uses one tool for a narrow task, while most of its operations remain unchanged. The government survey also found that most businesses using AI report significant human oversight, suggesting that deployment is often assistive rather than fully autonomous.
What does AI mean for jobs?
Current data do not show a broad employment shock. In the ONS survey conducted in late December 2025, 4% of businesses already using AI reported reduced headcount as a result. That is a reported outcome among current users, not a forecast for the whole economy.
Recommended Free Tools
The combination of limited technology breadth, extensive human involvement and the small share reporting headcount reductions is more consistent with task-level augmentation and controlled pilots than with wholesale replacement. Effects can still differ sharply by occupation: administrative, marketing and IT tasks are among the most exposed to workflow changes, while the business may retain people to review outputs, handle exceptions and manage accountability.
Who is responsible when business AI goes wrong?
Responsibility is usually distributed, but distribution should not mean ambiguity. In the Bank of England’s 2024 survey of financial-services firms, 72% said executive leadership was accountable for AI use cases, and 84% had an accountable person for their overall AI framework. Most firms, however, reported three or more accountable persons or bodies.
Rank #4
That pattern describes a governance design choice rather than proof that nobody is in charge. A workable model separates framework ownership from individual-use-case approval and records who can stop a system when controls fail.
Questions a sound accountability model should answer
- Who owns the AI framework? Name the senior role or committee responsible for policy, risk appetite and resourcing.
- Who approves each use case? Assign a business owner who understands the process, data and customer impact, rather than leaving approval to a general technology team.
- How is human oversight documented? Define when a person must review, override or escalate an output, and keep an audit trail.
- How are security, privacy and bias risks controlled? Record data permissions, testing, monitoring, incident response and retirement criteria.
- Is accountability centralised or distributed? A central committee can set consistent controls; operational owners can make decisions close to the work. The arrangement must still identify one escalation route and a final decision-maker.
Are British businesses ready for wider AI use?
Readiness is mixed. The adoption numbers show that many organisations are still deciding whether AI belongs in their operations, while the growth among larger firms indicates that capability and resources matter. Even among adopters, extensive use remains unusual and the typical business has added only a fraction of another AI technology since late 2023.
Businesses appear more prepared when they combine a limited, measurable use case with strategy, training and oversight. A pilot that has no named owner, data controls or stop conditions is not evidence of organisational readiness; it is simply evidence that a tool has been switched on.
How to judge an AI programme inside a UK business
- Define the outcome. State the process being improved, the measure of success and the people affected.
- Classify the risk. Identify whether errors could affect money, safety, legal rights, confidential data or reputation.
- Assign ownership before launch. Record the executive accountable for the framework and the operational owner for the use case.
- Set human checkpoints. Specify which outputs require review, what evidence the reviewer needs and who can override the system.
- Monitor and report. Track accuracy, incidents, complaints, security events and unintended effects rather than adoption alone.
- Review the decision. Expand only when controls work in practice; otherwise restrict, redesign or retire the use case.
What the numbers do—and do not—prove
The surveys establish a strong direction of travel: AI use has risen, larger and information-intensive businesses lead, and more firms are planning adoption. They do not establish that 35% of every UK business uses AI, because ONS and government studies cover different business sizes, samples and fieldwork periods. Nor do they show that adoption automatically produces productivity gains or widespread job losses.
The more defensible conclusion is narrower and more useful: British businesses are increasingly willing to deploy AI, but most deployments remain limited. The central business question is shifting from whether to try AI to whether the organisation can assign clear responsibility and prove that its controls match the consequences of using it.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
