AI Statistics UK 2026: Adoption, Productivity, Jobs & Business Data
Posted On: August 17, 2026
AI Workforce
Written by Luca Controlo, AI Adoption and Marketing Automation Lead at AI Workforce
Reviewed by Rodi Taze, Co-Founder of AI Workforce · Last updated: 17 August 2026
Every part of the UK economy is now producing its own AI numbers, and they do not always agree. The Office for National Statistics, the British Chambers of Commerce, Stanford's AI Index, the CIPD and the UK government are all measuring adoption, productivity, jobs and investment at the same time, using different samples, different definitions of "using AI" and different time periods. This guide brings together the most reliable UK statistics for 2026, with the original publisher, publication date, data period and a direct link for every figure, so each number can be checked rather than taken on trust.
Quick Answer: Around 35% of UK businesses with 10 or more employees reported using at least one AI technology in June 2026, according to the Office for National Statistics. A separate BCC survey, based predominantly on SMEs and using a broader measure, found 54% of responding firms were actively using AI. Adoption is rising, but the evidence consistently shows a gap between trying AI, embedding it in business processes, and achieving measurable productivity or revenue gains. The figures in this guide identify the source, publication date, data period, methodology and context behind each statistic.
Key UK AI Statistics for 2026
ONS: around 35% of UK businesses with 10 or more employees used at least one AI technology in June 2026, up from around 12% in late 2023.
ONS: adoption reaches 49% among businesses with 250 or more employees and 28% among businesses with 0 to 9 employees.
BCC: 54% of surveyed UK firms, around 94% of them SMEs, said they were actively using AI in 2026, up from 35% in 2025.
Salesforce (vendor-commissioned): 90% of UK sales organisations report some AI use, 54% specifically for prospecting.
PwC: UK AI-skilled wage premium reached 34.2% in 2025, up from 11% in 2024.
CIPD: 17% of employers who knew AI was used in their organisation expected it to shrink headcount over the following year.
Stanford AI Index: global organisational AI adoption reached 88% in 2025; 70% used generative AI in at least one business function.
UK government: £2 billion allocated to AI from 2026-27 to 2029-30 in Spending Review 2025, including a £1.6 billion AI strategy commitment for UKRI announced in February 2026.
How to Read and Compare These Statistics
Every principal statistic in this guide is presented with the original publisher, publication date, data period, source link and relevant context. Supporting figures are grouped under the same source when they come from the same study. This matters because UK AI statistics vary enormously depending on definition, sample and measurement date. A figure describing "any AI use by any UK firm regardless of size" will differ sharply from one measuring "a defined AI technology used by businesses with 10 or more employees," even when both are accurate for what they measure.
Three distinctions are worth holding in mind throughout. First, UK-specific figures should not be directly compared with global figures unless explicitly stated, since population, business mix and survey method differ by country. Second, self-reported estimates from workers or business owners are not the same as independently observed outcomes. Third, forecasts and projections describe a possible future, not a current state, and should be labelled as such.
UK AI Adoption
For a fuller treatment of adoption trends specifically, see our dedicated AI Adoption 2026 guide.
Around 35% of UK businesses with 10 or more employees reported using at least one AI technology in June 2026.
Context: This measure covers businesses with 10 or more employees only and records use of at least one specified AI technology, such as machine learning or natural language processing. It is up from around 12% in late 2023, just under three times the initial level. It should not be treated as an estimate covering every UK sole trader or microbusiness, which the ONS reports separately. |
Source: ONS, Artificial intelligence in UK businesses: 2023 to 2026, published 20 July 2026. Data period: late 2023 and 5 to 28 June 2026.
Adoption varies sharply by business size in the same ONS data. Businesses with 250 or more employees reported 49% adoption, while businesses with 0 to 9 employees reported 28% (ONS, 20 July 2026, data period 5 to 28 June 2026, ons.gov.uk). Context: the ONS headline chart does not publish separate percentages for the 10 to 49 or 50 to 249 employee bands, so those figures should not be quoted until the full ONS data table is checked directly.
In separate BCC research, 54% of surveyed UK firms said they were actively using AI in 2026, up from 35% in 2025.
Context: This is a survey result from a sample dominated by SMEs, not an official population estimate, and it uses a broader measure of "active AI use" than the ONS's technology-specific definition, which explains the gap with the ONS figure above. The same research found AI adoption in 2024 at 25% and in 2023 at 23%. |
Source: ONS, Artificial intelligence in UK businesses: 2023 to 2026, published 20 July 2026. Data period: 5 to 28 June 2026.
Depth of adoption remains limited even where headline adoption is high. Only 10% of ONS-surveyed AI-adopting businesses with 10 or more employees report "extensive" AI use, with the figure highest among the smallest adopting businesses at 17% and lowest among the largest at 9% (ONS, 20 July 2026, ons.gov.uk). Context: This suggests that broad adoption figures like the ONS's 35% or the BCC's 54% describe how many businesses have tried AI, not how deeply it is embedded in day-to-day operations.
AI Adoption by Industry
Information and communication businesses report the highest AI adoption of any UK industry, at 58%, compared with 13% in construction.
Context: This roughly 4.5 times gap between the highest and lowest adopting industries has persisted since ONS began tracking the question in 2023, and reflects both digital maturity and how easily a sector's core tasks map onto existing AI tools. |
Source: ONS, Artificial intelligence in UK businesses: 2023 to 2026, published 20 July 2026. Data period: 5 to 28 June 2026.
AI Technologies Used by UK Businesses: LLMs 18%; visual content creation 16%; ML data processing 12%; ML image processing 6%; Robotics 2%
Source: ONS, Artificial intelligence in UK businesses: 2023 to 2026, published 20 July 2026. Data period: 5 to 28 June 2026.
Among UK businesses with 10 or more employees, large language models are the most commonly reported AI technology, used by 18%, followed by visual-content creation at 16%, machine-learning data processing at 12%, machine-learning image processing at 6% and robotics at 2% (ONS, 20 July 2026, ons.gov.uk). Context: text and image generation require the least integration work of any AI application, which likely explains why they lead adoption even in sectors that have not adopted AI more broadly. The average adopting business now uses 1.6 AI technologies, up from around 1.4 in 2023.
Not sure where your business sits against these figures?
See which processes are genuinely ready for AI before you commit budget.
AI Productivity
For a fuller treatment of UK productivity trends specifically, see our dedicated AI Productivity guide.
In a vendor-sponsored survey of 1,500 UK digital workers, respondents estimated AI saved them 12 hours a week, while 6.3 hours were spent supplying context, supervising outputs and correcting results.
Context: Both figures are self-reported estimates from a vendor-commissioned study, not independently observed time-and-motion data, and the two estimates may overlap or use different question structures. This guide does not treat the net time saving as a confirmed 5.7 hours, since the original methodology has not been checked to confirm the two figures are directly subtractable. |
The same survey found 90% of respondents are now required to use AI in their role, and 80% use multiple AI tools weekly, yet only 18% agreed AI had significantly improved their organisation's overall performance (Work AI Index: UK 2026, uctoday.com). Context: This suggests individual-level time savings are not automatically translating into organisation-wide performance gains.
In BCC research, UK SMEs already using AI reported a net productivity expectation of +71 percentage points.
Context: A net productivity expectation balances the share of firms expecting improved productivity against those expecting it to worsen; it is a forward-looking sentiment measure, not a measured output gain, and should not be restated as "75% of businesses report a productivity gain." |
Improving business operations is the most commonly reported purpose for using AI among UK adopters, cited by close to 60% of businesses across the size bands measured and by more than 60% of larger businesses. Among businesses using AI to improve operations, 63% reported no change in workforce headcount, 6% reported a decrease and 1% reported an increase.
Context: Improving business operations may include administrative and back-office processes, but the ONS does not publish a separate representative measure of hours saved specifically on administration. This guide therefore does not convert this adoption figure into an administrative time-saving claim. |
AI in Sales
90% of UK sales organisations report using some form of AI for tasks such as prospecting, forecasting, lead scoring or drafting emails, and 54% are specifically using AI for prospecting, with a further 37% planning to.
Context: This is vendor-commissioned research with a disclosed UK sample and published methodology, not an independent government or academic survey. AI agent use specifically is less common: 46% of UK sellers report having used agents, and UK sellers expect agents to eventually cut prospect research and email drafting time by 38% each, an expectation rather than a measured outcome. |
This mirrors the wider adoption-versus-depth pattern seen throughout this guide: broad experimentation with AI sales tools is common in the UK, but sustained, measured production use, with proper CRM logging, escalation rules and human review, is harder to verify independently. For a fuller treatment, see our dedicated AI Sales Statistics 2026 guide.
AI in Customer Service
Globally, 74% of consumers say that AI has led them to expect customer service to be available 24 hours a day.
Context: This measures customer expectations, not the proportion of UK businesses currently offering AI-powered support or the performance of AI receptionists, chatbots or voice agents. |
This review did not identify a representative UK-wide benchmark measuring chatbot, voice AI or AI-receptionist adoption. The absence of such a benchmark should not be interpreted as evidence that adoption is either widespread or limited. AI receptionist and voice AI systems are increasingly marketed for after-hours and overflow coverage, covered in more operational detail in our guide to AI receptionist pricing in the UK. For a fuller, separately sourced treatment, see our dedicated AI Customer Service Statistics 2026 guide.
AI and the UK Workforce
17% of employers who knew AI was being used in their organisation expected it to reduce workforce size over the following year.
Context: This is an employer expectation about the next 12 months, not an observed redundancy rate, and it was published in November 2025. The risk was concentrated in large private sector firms, where 26% of employers expected headcount to fall, compared with 17% across the private sector overall and 20% in the public sector. |
Government-supported modelling estimated that jobs directly involving some AI activity could increase from around 158,000 in 2024 to 3.9 million by 2035.
Context: This is a projection over an 11-year horizon, not a current employment count. It should not be interpreted as 3.7 million entirely new AI jobs; it represents an upper-bound estimate combining transformed roles, net employment growth and replacement demand. |
AI and the UK Workforce: Three Different Measures: 17% next 12 months; employers expecting headcount fall (CIPD); 95% past year SMEs reporting no headcount impact (BCC); 3.9m government projection; jobs involving AI activity (2024 to 2035)
Sources: CIPD, Autumn 2025 Labour Market Outlook (10 Nov 2025); BCC and Atos, Future of Work report (18 March 2026); DSIT/DCMS labour market and skills projections (28 January 2026). These measures are not directly comparable and are shown separately.
Separately, BCC research found 95% of SMEs already using AI reported no impact on workforce size over the past year, and 86% reported no change to job roles (BCC and Atos, 18 March 2026, britishchambers.org.uk). Context: read together, the CIPD and BCC findings are not contradictory: the CIPD measures employer expectations weighted toward larger private firms, while the BCC's SME-heavy sample reports minimal headcount impact so far.
AI Skills and Training
Job postings for specialist AI roles in the UK rose 61% year on year, from around 112,000 to around 180,000 in 2025, and the average wage premium for AI-skilled workers nearly tripled to 34.2%, up from 11% in 2024.
Context: This measures job-posting and wage data, not a direct skills-gap survey, and the wage premium varies sharply by sector, peaking at 64% in consumer markets and 12% in government and the public sector. Specialist AI roles remain a small share of the overall UK job market, at 2.2% of postings. |
Separately, the ONS's own business survey found that 62% of AI-adopting UK businesses reporting a lack of AI expertise as a barrier had responded by training or retraining staff, compared with 26% of businesses reporting no barriers, and around 40% of medium-to-large adopting businesses report integrating AI skills through training, though only 11% report more than half their workforce has received AI-related training so far (ONS, 20 July 2026, ons.gov.uk).
UK AI Investment
The UK's Spending Review 2025 allocated £2 billion to AI from 2026-27 to 2029-30.
Context: This is a multi-year public funding commitment, not money already spent, and the settlement dates from June 2025 even though the funding period begins in April 2026. |
UK Research and Innovation committed £1.6 billion directly targeted at the AI sector between 2026 and 2030, its biggest single investment area for the period.
Context: The UKRI announcement states this £1.6 billion is drawn from the recent Spending Review settlement above. This guide does not add the £2 billion and £1.6 billion figures together, since the government has not confirmed the two amounts are fully non-overlapping. |
UK Government AI Funding Timeline: Spending Review 2025 (published June 2025) £2bn allocated to AI 2026-27 to 2029- 30; UKRI AI Strategy announced 19 Feb 2026: £1.6bn, drawn from the Spending Review settlement (not additional funding), same settlement
Sources: HM Government, Spending Review 2025 (published June 2025); UKRI and DSIT, AI Strategy for UK Research and Innovation (announced 19 February 2026).
No single representative measure of private UK business AI spending for 2026 was identified. Government announcements frequently combine multi-year commitments, infrastructure projects and anticipated investment, so they should not be treated as money already deployed. The most recent formal sector-level study, DSIT's AI Sector Study, was published in October 2024 using 2023 data, and is too dated to represent current 2026 private investment levels; this guide has excluded it rather than presenting outdated figures as current.
What Does Global AI Adoption Look Like, for Comparison?
Global organisational AI adoption reached 88% in 2025, and 70% of surveyed organisations used generative AI in at least one business function.
Context: These are global figures describing organisations surveyed internationally in 2025, not UK-specific figures, and should not be directly compared with the UK's ONS or BCC adoption figures above, which use different definitions, samples and countries. |
Stanford's Index separately reports that generative AI reached an estimated 53% of global population adoption within roughly three years of ChatGPT's public launch (Stanford HAI, April 2026, hai.stanford.edu). Context: this is a population-level adoption estimate, distinct from both organisational adoption and any UK business-level statistic in this guide.
What Should UK Businesses Take From the 2026 Data?
Read together, the verified 2026 statistics tell a consistent story: adoption is genuinely rising in the UK, from around 12% of businesses with 10 or more employees in late 2023 to around 35% in June 2026, but depth of adoption, measured productivity gains and confirmed workforce impact all lag well behind headline usage figures. Employer expectations about job losses and researcher projections about job growth can both be true at the same time, since they describe different time horizons, different firm sizes and different types of roles.
A UK business deciding what to do with AI in 2026 is better served by asking where a specific, well-defined workflow can be improved than by chasing adoption for its own sake, and by treating vendor-sourced statistics, including several figures excluded from this guide, with appropriate caution until they can be traced to a primary source. That is also where an AI agent cost review and the requirements under UK GDPR become relevant before committing budget to a specific workflow.
Methodology and Source Standards
Every principal statistic in this guide is presented with the original publisher, publication date, data period, source link and relevant context. Supporting figures are grouped under the same source where they come from the same study. Where a figure could not be traced to an identifiable primary publisher with disclosed methodology, it has been excluded rather than repeated from a secondary aggregator. Vendor-commissioned research is identified as such. UK-specific and global figures are kept clearly separated. Forecasts and projections are labelled as forecasts, not treated as current measurements.
Frequently Asked Questions
What percentage of UK businesses use AI in 2026?
According to the ONS, around 35% of UK businesses with 10 or more employees reported using at least one AI technology in June 2026, up from around 12% in late 2023. A separate BCC survey, based predominantly on SMEs and using a broader measure of active AI use, found 54% of responding firms were actively using AI.
Is AI actually saving UK workers time?
In a vendor-sponsored survey of 1,500 UK digital workers, respondents estimated AI saved them around 12 hours a week, while around 6.3 hours were spent supplying context, supervising outputs and correcting results. Both figures are self-reported estimates from a single commissioned study, not independently measured outcomes.
Will AI reduce jobs in the UK?
The evidence is mixed and depends on time horizon and firm size. The CIPD found 17% of employers who knew AI was in use expected it to shrink their workforce over the following year, with risk concentrated in large private firms, while a separate government skills projection published in January 2026 expects jobs directly involving AI activity to rise from around 158,000 in 2024 to 3.9 million by 2035.
Key Takeaways
ONS: around 35% of UK businesses with 10 or more employees used at least one AI technology in June 2026, up from around 12% in late 2023.
Adoption varies sharply by size and industry: 49% among businesses with 250 or more employees versus 28% among businesses with 0 to 9 employees; 58% in information and communication versus 13% in construction.
BCC: 54% of surveyed UK firms, predominantly SMEs, reported active AI use in 2026, but only 10% of ONS-tracked adopters report "extensive" use.
Productivity and time-saving figures circulating for 2026 are mostly self-reported and vendor-commissioned; treat headline hour-savings and productivity percentages as estimates, not measured outcomes.
The CIPD found 17% of employers expected AI to shrink headcount over the next year, while government projections expect AI-related jobs to grow to 3.9 million by 2035.
UK AI-skilled wage premiums nearly tripled to 34.2% in 2025, and specialist AI job postings rose 61% year on year, per PwC's 2026 AI Jobs Barometer.
90% of UK sales organisations report some AI use, and 74% of consumers expect round-the-clock customer support, per vendor-commissioned Salesforce and Zendesk surveys with disclosed samples and methodology.
UK government AI funding for 2026 to 2030 totals £2 billion from Spending Review 2025, with a related £1.6 billion UKRI commitment drawn from the same settlement, not an additional sum.
Full Source Register
Office for National Statistics, Artificial intelligence in UK businesses: 2023 to 2026, published 20 July 2026.
British Chambers of Commerce and Atos, Future of Work: AI in the Workplace Report, published 18 March 2026.
CIPD, Autumn 2025 Labour Market Outlook press release, published 10 November 2025.
PwC, 2026 AI Jobs Barometer press release, published 15 June 2026.
Salesforce, State of Sales 2026 (vendor-commissioned), published 3 February 2026.
Zendesk, CX Trends 2026 (vendor-commissioned, global evidence), published 2026.
DSIT and DCMS, AI Skills for Life and Work: Labour Market and Skills Projections, published 28 January 2026.
HM Government, Spending Review 2025, published June 2025.
UKRI and DSIT, AI Strategy for UK Research and Innovation, published 19 February 2026.
Stanford HAI, The 2026 AI Index Report, Economy chapter, published April 2026.
The Work AI Index: UK 2026, commissioned by Glean's Work AI Institute, reported in 2026.
Related Reading
Want help selecting and measuring a first AI workflow?
Get a tailored AI readiness review for your team.
Frequently Asked Questions
Everything you need to know about this topic
Global organisational AI adoption reached 88% in 2025, and 70% of surveyed organisations used generative AI in at least one business function. SourceStanford HAI, The 2026 AI Index Report, Economy chapterPublishedApril 2026Data period2025NoteGlobal evidence; included for context.Source linkRead the Stanford AI Index economy chapter Context: These are global figures describing organisations surveyed internationally in 2025, not UK-specific figures, and should not be directly compared with the UK's ONS or BCC adoption figures above, which use different definitions, samples and countries. Stanford's Index separately reports that generative AI reached an estimated 53% of global population adoption within roughly three years of ChatGPT's public launch (Stanford HAI, April 2026, hai.stanford.edu). Context: this is a population-level adoption estimate, distinct from both organisational adoption and any UK business-level statistic in this guide.
Read together, the verified 2026 statistics tell a consistent story: adoption is genuinely rising in the UK, from around 12% of businesses with 10 or more employees in late 2023 to around 35% in June 2026, but depth of adoption, measured productivity gains and confirmed workforce impact all lag well behind headline usage figures. Employer expectations about job losses and researcher projections about job growth can both be true at the same time, since they describe different time horizons, different firm sizes and different types of role. A UK business deciding what to do with AI in 2026 is better served by asking where a specific, well-defined workflow can be improved than by chasing adoption for its own sake, and by treating vendor-sourced statistics, including several figures excluded from this guide, with appropriate caution until they can be traced to a primary source. That is also where an AI agent cost review and the requirements under UK GDPR become relevant before committing budget to a specific workflow.
According to the ONS, around 35% of UK businesses with 10 or more employees reported using at least one AI technology in June 2026, up from around 12% in late 2023. A separate BCC survey, based predominantly on SMEs and using a broader measure of active AI use, found 54% of responding firms were actively using AI.
In a vendor-sponsored survey of 1,500 UK digital workers, respondents estimated AI saved them around 12 hours a week, while around 6.3 hours were spent supplying context, supervising outputs and correcting results. Both figures are self-reported estimates from a single commissioned study, not independently measured outcomes.
The evidence is mixed and depends on time horizon and firm size. The CIPD found 17% of employers who knew AI was in use expected it to shrink their workforce over the following year, with risk concentrated in large private firms, while a separate government skills projection published in January 2026 expects jobs directly involving AI activity to rise from around 158,000 in 2024 to 3.9 million by 2035.
