Enterprise AI adoption: how a large company catches up on AI / Where you stand
How many companies use AI: the official figures by country, size and sector
The newest official US count comes from the Census Bureau's survey collected 5 to 18 October 2026: 24.6% of US businesses used artificial intelligence (AI) in the previous two weeks, and 43.6% of those with 250 or more employees. In the European Union, 19.95% of enterprises with 10 or more employees used AI in 2025, and 55.03% of large ones. In the United Kingdom, 38% of businesses with 10 or more employees used AI in September 2026. In Canada, 19.2% of businesses used AI in the 12 months before a survey run in April and May 2026. This page gives each figure with its question, its base and its date, and explains why the surveys disagree.
Published October 26, 2026. Editorial.
Key takeaways
- The US Census Bureau counted 24.6% of businesses using AI in the two weeks of 5 to 18 October 2026, up from 17.3% in the period of 17 to 30 November 2025.
- For US firms with 250 or more employees the figure was 43.6%, with a standard error of 1.51 points, and it moved by up to 7.5 points between neighbouring periods.
- Eurostat counted 55.03% of large EU enterprises using AI in 2025, against 17% of small ones, with Denmark highest at 42.03% of all enterprises.
- Each survey asks a different question, so compare a company only with a figure that uses the same question, base and period.
- In the Census Bureau's AI supplement, 22.9% of respondents at firms with 250 or more employees did not know whether their firm used AI.
In the two weeks from 5 to 18 October 2026, 24.6% of US businesses said they had used artificial intelligence (AI) in one of their business functions [1]. The US Census Bureau published that figure on 22 October 2026. For businesses with 250 or more employees, the share was 43.6% [2]. In the European Union (EU), 55.03% of enterprises with 250 or more employees used AI in 2025 [10].
This page gives the newest official figure from each of four statistics offices, with the question each office asked, the businesses it counted and the dates. It then explains why private surveys give higher numbers. It is part of the guide on enterprise AI adoption.
The United States: the Census Bureau's survey
The Business Trends and Outlook Survey (BTOS) is the Census Bureau's survey of US businesses. By the Census Bureau's description, its sample of 1.2 million businesses is split into six panels. Data is collected every two weeks, and each panel is asked to report once every 12 weeks [5]. The AI question asks whether the business used AI in any of its business functions in the last two weeks.
A note on the name. Executive Order 14434, signed on 29 September 2026, tells US executive agencies to write "Super Intelligence" and "SI" in place of "Artificial Intelligence" and "AI" [4]. Because of that order, the Census Bureau now calls its AI figures "Super Intelligence" or "SI": its files say "As of October 22, 2026, data is published in compliance with Executive Order 14434" [1]. The survey question asked about "Artificial Intelligence (AI)" until the collection that began on 19 October 2026 [5]. Every figure on this page was collected before that date, so this page says AI.
By company size
The size file gives the share of businesses using AI for each size class, and the national file gives the share for all businesses. The four most recent periods read as follows [1][2].
| Employees | 5 to 18 Oct 2026 | Period before | Two periods before | Three periods before |
|---|---|---|---|---|
| 1 to 4 | 24.3% | 23.9% | 23.8% | 23.6% |
| 100 to 249 | 35.3% | 36.4% | 34.1% | 34.5% |
| 250 or more | 43.6% | 36.5% | 44.0% | 38.4% |
| All businesses | 24.6% | 23.8% | 23.8% | 23.2% |
The large-firm figure moves more than the others. A standard error measures how far a survey estimate could be from the true figure because only a sample of businesses answered. The standard error for firms with 250 or more employees in the latest period is 1.51 percentage points, against 0.27 points for all businesses [2][1]. Between neighbouring periods, the large-firm figure moved by up to 7.5 points. Reveneau's view is that a board should read the large-firm figure as a range across several periods, and should not read a change between two periods as a trend.
In the same October period, 28.6% of all businesses said they expected to use AI in the next six months [1].
Two breaks in the series
The question has changed twice. On 17 November 2025 the Census Bureau changed the wording from AI use "in producing goods or services" to AI use in any of the business's functions. The level jumped, so the Bureau started a new series: "Due to a level shift observed in conjunction with the new question wording, the decision was made to create a new time series for the AI questions, beginning with data released on December 4, 2025" [6]. A figure from before November 2025 cannot be compared with a later one.
Within the new series, the share of all businesses using AI rose from 17.3% in the period collected 17 to 30 November 2025 to 24.6% in October 2026 [1]. A Census Bureau story of 26 May 2026 said that from December 2025 to May 2026 "overall AI usage hovered between 17% and 20%". The same story gave an example by size: "37% of firms with at least 250 employees reported using AI in their business operations" [7].
The second change came with the executive order. The data page says the wording was updated because of Executive Order 14434 "Beginning with the October 19 2026 collection" [5]. The period of 5 to 18 October 2026 is the last one collected before that change. The next releases may show another break, so check the Census notes before comparing them with the figures on this page.
By sector
The sector file uses industry codes. For the three codes checked against the Census Bureau's own story, the shares for 5 to 18 October 2026 were: information, 46.4%; finance and insurance, 40.5%; retail trade, 18.5% [3]. The Census story gave 39.7% for information and 33.9% for finance and insurance in the period ending 3 May 2026, and put retail trade at 14% in rounded terms [7]. A bank or insurer should also read the guide on AI in banking.
A second Census question shows a gap inside large firms
The Census Bureau also ran a longer AI supplement, collected 17 November 2025 to 8 February 2026 [8]. In it, 30.6% of respondents at firms with 250 or more employees said the firm had used AI in the last two weeks, and 22.9% said they did not know [8]. The supplement ran over different weeks from the October figure, so the two shares should not be compared. Reveneau reads the 22.9% as a sign that, in a large company, the person who answers a survey cannot always see all the AI the company uses. Managers inside the company can have the same limited view, which is why the first 90 days checklist starts with an inventory.
Other US surveys disagree
A Federal Reserve Board note of 3 April 2026 cites earlier work by Crane, Green and Soto (2025), who examined 16 surveys of AI use in the US. As of mid-2024, the surveys' estimates of work-related AI adoption ranged from 5 to 40 percent [9]. The note gives several reasons the estimates differ: which sizes of firm a survey samples, how the question is framed, whether the person answering knows the firm's AI use, and pressure on senior leaders to report AI use [9].
The European Union: Eurostat
Eurostat, the EU's statistics office, counts enterprises with 10 or more employees and self-employed persons. Its 2025 figures come from surveys held by national statistics offices in the first months of 2025. In 2025 it surveyed 157,000 of the 1.53 million enterprises in the EU [10]. Its main figure is the share that used at least one AI technology.
| Enterprises | Share using AI in 2025 | Source |
|---|---|---|
| All with 10 or more employees | 19.95% | [10] |
| Small (10 to 49 employees) | 17% | [10] |
| Medium (50 to 249 employees) | 30.36% | [10] |
| Large (250 or more employees) | 55.03% | [10] |
| All, in 2024 | 13.48% | [11] |
| Large, in 2021 | 28.41% | [11] |
The share for all enterprises rose by 6.47 percentage points from 2024 to 2025 [10]. The share for large enterprises went from 28.41% in 2021 to 55.03% in 2025 [11].
The highest country share is eight times the lowest. "The highest share was recorded in Denmark (42.03%), followed by Finland (37.82%) and Sweden (35.04%), while the lowest shares were recorded in Romania (5.21%), Poland (8.36%) and Bulgaria (8.55%)" [10]. Sectors differ too. The information and communication sector led at 62.52%, followed by professional, scientific and technical activities at 40.43%, and in every other sector the share was below 25% [10]. Construction was lowest, at 10.79% [10].
Finance and insurance is outside the list of sectors Eurostat covers [10], so a bank or insurer in the EU has no Eurostat figure for its own sector. The EU's target is that "By 2030, three out of four EU companies should use cloud computing services, big data or artificial intelligence" [10].
The United Kingdom: the Office for National Statistics
The Office for National Statistics (ONS) asks businesses about AI in its Business Insights and Conditions Survey (BICS). In the wave that ran from 21 September to 4 October 2026, 30% of all UK businesses and 38% of businesses with 10 or more employees used at least one AI technology [12]. The 38% is "up 3 percentage points from June 2026 and 26 percentage points from September 2023" [12]. The information and communication industry was highest, at 60% [12].
The September bulletin gives 51% for the largest businesses, but it labels that group "250 or more employees" in one place and "100 or more employees" in another [12]. This page uses the June 2026 figure instead: 49% of UK businesses with 250 or more employees used AI, against 28% of those with 0 to 9 employees [13].
Two further ONS points matter to a large company. Among UK businesses with 10 or more employees that use AI, "only 10% report using AI extensively", from a new question in June 2026 that the ONS asks readers to treat as early indications [13]. And BICS leaves out some industries, including "public administration and defence, public provision of education and health, and finance and insurance" [13]. Banks, insurers and public bodies are outside the UK figure.
Canada: Statistics Canada
Statistics Canada asks whether a business used AI to produce goods or deliver services in the 12 months before the survey. The survey ran from 1 April to 6 May 2026, with 9,251 responses from a sample of 21,105 [14].
| Group | Share using AI in the last 12 months | Source |
|---|---|---|
| All businesses, 2026 | 19.2% | [14] |
| All businesses, 2025 | 12.2% | [14] |
| All businesses, 2024 | 6.1% | [14] |
| Businesses with 100 or more employees | 27.8% | [14] |
| Information and cultural industries | 42.3% | [14] |
| Finance and insurance | 40.4% | [14] |
Statistics Canada also reports that 40.0% of businesses "indicated the use of AI is not relevant to the business" [14].
Private counts are higher
The 2026 AI Index from Stanford University's Institute for Human-Centered AI says organisational AI adoption rose "up to 88% of surveyed organizations" in 2025, and that 70% used generative AI in at least one business function [15]. Generative AI is AI that creates text, images or code. The chapter page does not name the survey behind the 88%, so this guide reports it as the AI Index's figure only.
The ONS explains the gap: "Lower estimates tend to capture direct, bottom-up use (for example, by workers or specific business functions). Higher estimates reflect a broader, top-down view of whether firms use any AI technology across their operations" [13]. Workers also report more use than firms. In ONS surveys of May and June 2026, 55% of employees in Great Britain said they use AI for work or education, while 35% of UK businesses with 10 or more employees reported using at least one AI technology [13].
| Source | What it asks | Figure | Period |
|---|---|---|---|
| Census Bureau BTOS | Any AI use in the last two weeks, all businesses in the survey | 24.6% | 5 to 18 October 2026 [1] |
| Eurostat | At least one AI technology, enterprises with 10 or more employees | 19.95% | 2025 [10] |
| ONS BICS | At least one AI technology, businesses with 10 or more employees | 38% | September 2026 [12] |
| Statistics Canada | AI to produce goods or deliver services, last 12 months | 19.2% | Second quarter of 2026 [14] |
| Stanford AI Index | Organisational AI adoption, surveyed organisations | 88% | 2025 [15] |
What a large company should do with these figures
Reveneau's view is that the official figures show how a company compares with others of its size and sector. To learn what its own staff do with AI, a company needs its own inventory of the AI it uses, by team and by tool, with a named owner for each. The page AI training for staff covers what staff need to know, and choosing your first AI use cases covers what to build first. A fund looking across several companies should read AI adoption in portfolio companies.
Reveneau is an AI software development consultancy with offices in San Francisco and Vancouver. Reveneau says AI writes all of its code and that every change must pass an eval suite written from the specification before release. To talk about a project, see Reveneau's services or contact Reveneau.
Common questions
What share of US businesses used AI in October 2026?
In the US Census Bureau's survey collected 5 to 18 October 2026, 24.6% of US businesses said they had used AI in a business function in the previous two weeks. The figure was published on 22 October 2026, with a standard error of 0.27 points. In the same period, 28.6% of businesses said they expected to use AI in the next six months.
How reliable is the Census AI figure for firms with 250 or more employees?
The Census AI figure for firms with 250 or more employees is less precise than the national figure. Its standard error in the 5 to 18 October 2026 period was 1.51 points, against 0.27 for all businesses. The large-firm figure read 43.6%, after 36.5%, 44.0% and 38.4% in the three periods before. Read it as a range across several periods.
Can Census AI figures from before November 2025 be compared with newer ones?
Census AI figures from before November 2025 cannot be compared with newer ones. On 17 November 2025 the Census Bureau changed the question from AI use in producing goods or services to AI use in any business function. The level jumped, so the Bureau started a new time series with data released on 4 December 2025. The wording changed again from the 19 October 2026 collection.
Which US industries use AI the most?
In the Census Bureau's survey for 5 to 18 October 2026, 46.4% of businesses in the information sector used AI, and 40.5% in finance and insurance. Retail trade was at 18.5%. A Census story gave 39.7% for information, 33.9% for finance and insurance and a rounded 14% for retail in the period ending 3 May 2026, so all three were higher in October.
Which EU countries have the highest share of companies using AI?
Denmark had the highest share of enterprises using AI in the EU in 2025, at 42.03%, followed by Finland at 37.82% and Sweden at 35.04%, according to Eurostat. The lowest shares were in Romania at 5.21%, Poland at 8.36% and Bulgaria at 8.55%. The figures cover enterprises with 10 or more employees.
Does Eurostat's AI figure include banks and insurers?
Eurostat's AI figure does not include banks and insurers. Its survey covers manufacturing, energy, water supply, construction, trade and other listed sectors, and finance and insurance is outside that list. A bank or insurer in the EU therefore has no Eurostat figure for its own sector. In 2025 Eurostat surveyed 157,000 of the 1.53 million enterprises in the EU.
How many UK businesses use AI?
In the Office for National Statistics survey that ran from 21 September to 4 October 2026, 30% of all UK businesses and 38% of businesses with 10 or more employees used at least one AI technology. The 38% was up 26 percentage points from September 2023. The survey leaves out finance and insurance, public administration and public education and health.
How many Canadian businesses use AI?
Statistics Canada reports that 19.2% of Canadian businesses used AI to produce goods or deliver services in the 12 months before its survey of 1 April to 6 May 2026. The share was 12.2% in 2025 and 6.1% in 2024. Among businesses with 100 or more employees, 27.8% used AI, and 40.0% of all businesses said AI was not relevant to them.
How many respondents at large US firms did not know whether their firm used AI?
In the US Census Bureau's AI supplement, collected 17 November 2025 to 8 February 2026, 22.9% of respondents at firms with 250 or more employees said they did not know whether the firm had used AI in the last two weeks. In the same supplement, 30.6% said yes. The answer suggests that in a large firm, the person filling in a survey may not see all the AI in use.
Why does the Stanford AI Index report that 88% of organizations use AI?
The Stanford AI Index 2026 reports that 88% of surveyed organizations used AI in 2025. The chapter page does not name the survey behind the figure, so its question and sample cannot be compared with the official surveys. The UK Office for National Statistics explains that top-down questions about any AI use in a firm give higher estimates than questions about use by workers or specific functions.
Do more workers use AI than their employers report?
More workers report using AI than businesses do, according to the UK Office for National Statistics. In its article of 20 July 2026, 55% of employees in Great Britain said they use AI for work or education, while 35% of UK businesses with 10 or more employees reported using at least one AI technology. Employees can use AI tools that their employer has not adopted or recorded.
References
- [1] U.S. Census Bureau, Business Trends and Outlook Survey, National file (period collected 5 to 18 October 2026, published 22 October 2026): 24.6% used AI in the last two weeks, standard error 0.27, after 23.8%, 23.8% and 23.2%; 28.6% expect to use AI in the next six months; 17.3% in the period of 17 to 30 November 2025; data published in compliance with Executive Order 14434 as of 22 October 2026.
- [2] U.S. Census Bureau, Business Trends and Outlook Survey, Employment Size Class file (same period): 250 or more employees 43.6%, standard error 1.51, after 36.5%, 44.0% and 38.4%; 100 to 249 employees 35.3%; 1 to 4 employees 24.3%.
- [3] U.S. Census Bureau, Business Trends and Outlook Survey, Sector file (same period): information 46.4%, finance and insurance 40.5%, retail trade 18.5%.
- [4] The White House, Inaugurating the Era of Super Intelligence, Executive Order 14434 (29 September 2026): the executive branch shall use the terms Super Intelligence and SI in place of Artificial Intelligence and AI.
- [5] U.S. Census Bureau, Business Trends and Outlook Survey data page (read 26 October 2026): label Super Intelligence or SI (collected as Artificial Intelligence or AI); sample of 1.2 million businesses in 6 panels, each reporting once every 12 weeks, with collection every 2 weeks; wording updated because of Executive Order 14434 from the 19 October 2026 collection.
- [6] U.S. Census Bureau, BTOS AI Core Question Updates (3 December 2025): new wording and a new time series from data released on 4 December 2025.
- [7] U.S. Census Bureau, America Counts, Large Firms With at Least 20 Employees Biggest AI Users, by Grundy, Breaux and Khatiwoda (26 May 2026): use between 17% and 20% from December 2025 to May 2026; 37% of firms with at least 250 employees; information 39.7%, finance and insurance 33.9% and retail trade 14% (the story's rounded figure) as of 3 May 2026.
- [8] U.S. Census Bureau, BTOS AI Supplement Table 2026 (collected 17 November 2025 to 8 February 2026, published 23 April 2026): firms with 250 or more employees, 30.6% yes and 22.9% do not know.
- [9] Board of Governors of the Federal Reserve System, Monitoring AI Adoption in the US Economy, FEDS Notes (3 April 2026): cites Crane, Green and Soto (2025) on 16 surveys with estimates of work-related adoption from 5 to 40 percent as of mid-2024; reasons the estimates differ.
- [10] Eurostat, Use of artificial intelligence in enterprises, Statistics Explained (data extracted December 2025): 19.95% in 2025; by size, country and sector; 157,000 of 1.53 million enterprises surveyed; the 2030 target.
- [11] Eurostat, dataset isoc_eb_ai, Artificial intelligence by size class of enterprise (updated 15 June 2026): 13.48% of enterprises in 2024; 28.41% of large enterprises in 2021.
- [12] Office for National Statistics, Business insights and impact on the UK economy: 8 October 2026 (Wave 165, 21 September to 4 October 2026): 30% of all businesses and 38% of those with 10 or more employees used AI; information and communication 60%; 51% for the largest businesses, labelled two ways.
- [13] Office for National Statistics, Artificial intelligence in UK businesses: 2023 to 2026 (20 July 2026): 49% of businesses with 250 or more employees in June 2026; 10% use AI extensively; excluded industries; why estimates differ; 55% of employees in Great Britain against 35% of businesses.
- [14] Statistics Canada, Analysis on artificial intelligence use by businesses in Canada, second quarter of 2026, by Do, Sood and Johnston (11 June 2026): 19.2% of businesses; 27.8% with 100 or more employees; sector shares; 40.0% say AI is not relevant; sample of 21,105.
- [15] Stanford HAI, The 2026 AI Index Report, Economy chapter page: 88% of surveyed organizations used AI in 2025; 70% used generative AI in at least one business function.
Related reading
What the October 2026 Census AI figures tell a large company deciding how fast to move
The US Census Bureau counts 24.6% of businesses and 43.6% of firms with 250 or more employees as AI users in early October 2026. The figures count use in a two-week window. Our view is that a large company should set its pace from tests on its own tasks.
Wonderful went from zero to a $5 billion valuation in 20 months. Here is what it proves about enterprise AI
The fastest-growing enterprise AI company of the past two years sells deployment, and its investors paid for the final work of getting AI into production, which tells you where the value in AI now is.
Who pays for the last mile in enterprise AI?
Wonderful's 52 percent gross margin puts a price on getting AI into production for the first time at scale, and there are only three ways that cost ever gets paid.
Ask an AI vendor for its time to production, and skip the demo
Each of Wonderful's customer stories starts with the time it took to reach production. That time is the better proof, because it measures the vendor, while a demo only measures the model.