AI in the Workplace Statistics 2026: 52% Use It, 15% Daily, 25% Have a Plan

Ask how many people use AI at work and you can find a credible-looking answer anywhere between 21% and 78%. Both ends are real numbers from serious institutions. They measure different things: workers or firms, any use or regular use, generative AI or AI of any kind, a head count or a weighting by employment. This page puts the main primary series side by side, with the question each one actually asks, so you can cite the right number for the claim you are making.

The backbone is Gallup's quarterly workforce study, which has asked the same frequency question of roughly 19,000–24,000 employed US adults every quarter since 2023. Around it sit Pew Research Center's worker surveys, the US Census Bureau's Business Trends and Outlook Survey, the Federal Reserve Board's 2026 review of adoption data and the St. Louis Fed's time-savings estimates. Five metrics are our own derivations; each is labelled and its arithmetic shown.

AI at work benchmark cheat sheet

The figures most often needed in a board paper, an HR strategy deck or an article, with the period each one refers to. Full sources are in the methodology section.

1. How many employees use AI at work

Gallup asks employed US adults how often they use AI in their role, from "never" to "daily". It reports three cuts: total users (a few times a year or more), frequent users (a few times a week or more) and daily users. This is the longest consistent worker-level series available.

Sources: Gallup Workplace, "AI Use at Work Has Nearly Doubled in Two Years" (June 15, 2025), "AI Use at Work Rises" (December 14, 2025), "Frequent Use of AI in the Workplace Continued to Rise in Q4" (January 25, 2026), "Rising AI Adoption Spurs Workforce Changes" (April 12, 2026) and "Organizational AI Adoption Jumps Six Points" (July 20, 2026). Gallup publishes the Q2 2024 daily figure (4%) but not the matching total and frequent figures in these articles; dashes mark values not published. The Q4 2025 total is the 46% stated as "the previous quarter" in the Q1 2026 release.

Three things stand out. Total use more than doubled between 2023 and 2025, then the pace dropped sharply: +5 points from Q2 to Q3 2025, +1 point in Q4, +4 in Q1 2026 and +2 in Q2 2026. Gallup itself described the Q4 2025 total as flat. Meanwhile frequent and daily use kept rising every single quarter. And the first reading above 50% arrived only in Q1 2026, so "most workers use AI" became true by a margin of one to two points — within reach of the survey's own error band for subgroups.

2. Original metric: the daily-user share is the number to watch

Because total adoption has slowed, the more informative measure is how many of the people who use AI use it every day. We divide Gallup's daily share by its total share for each quarter where both are published.

In twelve months total AI use rose 12 points, from 40% to 52%. Over the same year the share of AI users who use it daily rose from one in five to nearly three in ten. The story of 2026 is depth, not reach.

The practical implication for employers: the workforce is splitting into a group for whom AI is part of every working day and a group for whom it is an occasional experiment or nothing at all. Measuring "adoption" as a single yes/no number hides that split. The rounding in Gallup's published percentages means each derived share carries roughly ±1 point of arithmetic uncertainty; the direction across five quarters is the finding.

3. The remote-capable and frontline divide

Gallup splits its sample by whether a job can be done remotely. The gap is large on every measure, and it grows as the measure gets more demanding.

Source: Gallup, "Frequent Use of AI in the Workplace Continued to Rise in Q4", January 25, 2026. Multipliers are our calculation. Gallup's June 2025 release adds that frequent use among production and frontline workers was 11% in 2023 and 9% in 2025 — essentially flat while white-collar frequent use reached 27%.

The intensity multiplier — 2.06× for any use, 2.35× for weekly use, 2.71× for daily use — is the clearest way to state the divide. Frontline workers are not simply later adopters; when they do use AI, they use it less often. Our frontline and deskless workforce statistics show the same pattern in recognition and communication: tools built for a desk reach the desk first.

Leaders, managers and individual contributors

Leaders are 1.73× as likely as individual contributors to use AI at all and 1.91× as likely to use it frequently (our ratios). A separate Gallup survey fielded January–March 2026 found regular (daily or weekly) use among 37% of managers against 25% of individual contributors, and among 40% of college graduates against 17% of workers without a degree. Pew finds the same education gradient: the share of workers with a bachelor's degree or more who do some of their work with AI rose from 20% to 28% between 2024 and 2025, against 13% to 16% for workers with some college or less.

4. AI use by industry

Two independent sources rank industries — Gallup by employee, the Census Bureau by firm. They agree on the top and the bottom.

Sources: Gallup Q4 2025 release (January 25, 2026); Gallup Global Indicator: Artificial Intelligence, figures as of May 2026; US Census Bureau, "Large Firms With at Least 20 Employees Biggest AI Users", May 26, 2026. Gallup's "technology" and the Census "Information" sector are close but not identical definitions. Retail daily use in Q4 2025 was 10% per Gallup.

By the Census measure, an information-sector firm is 2.8× as likely as a retailer to use AI (39.7% against about 14%). By Gallup's, a technology employee is 2.3× as likely as a retail employee to use it at all (77% against 33%). The Census Bureau also reports that use rose between December 2025 and May 2026 for firms with at least 20 employees but did not change significantly for firms with fewer than 20; fewer than 20% of firms with four or fewer employees use AI.

5. What employers are (and are not) doing

Source: Gallup quarterly releases. Gallup added the "don't know" option in Q3 2025, so this series is only comparable from that quarter onward; the 44% figure from Q2 2025 that still circulates was measured without it.

The six-point jump to 47% in Q2 2026 is the largest single-quarter move in the series. But integration has outrun direction. In May 2026 only 25% of employees said their organisation had communicated a clear plan for integrating AI; a year earlier it was 22%, alongside 30% reporting general guidelines or formal AI policies.

Manager behaviour is the lever Gallup singles out. Only 36% of employees in AI-integrating organisations strongly agree their manager supports the team's use of AI. Those who do are 1.7× as likely to use AI a few times a week or more, 7.4× as likely to strongly agree AI gives them more opportunity to do what they do best, and 8.7× as likely to strongly agree it has transformed how work gets done. For the wider evidence on why the manager is the multiplier, see our manager effectiveness statistics.

6. What employees use AI for

Source: Gallup, "Organizational AI Adoption Jumps Six Points", July 20, 2026.

The pattern is an inverse one: the most common uses (writing, search) report the lowest productivity payoff, while the less common technical uses (coding, automation, analytics) report the highest. Breadth matters even more than the specific task. Among users who apply AI to one or two purposes, 45% report a positive productivity effect; at three to four purposes it is 66%, at five to six 78%, and at seven or more 90%. Pew's October 2024 data on chatbot users points the same way: research (57%), editing (52%) and drafting (47%) were the leading uses, and 40% found chatbots extremely or very helpful for working faster against 29% for improving quality.

7. Productivity and time saved

What happens to those hours is the open question. Gallup's 2026 data suggests a meaningful share is being reabsorbed: 31% of AI users say their employer has asked them to take on more responsibility, rising to 37% among those who say AI makes them faster. That is the point where AI adoption becomes a reward and recognition question — the employees doing more because of a tool rarely see that reflected anywhere but their workload. Our employee productivity statistics cover the wider output picture.

8. Job security and worry

The job-elimination expectation held at 15% for three years while adoption doubled, then moved to 18% in early 2026. Adopting organisations report both more hiring and more reductions than non-adopters, which reads as churn rather than straightforward replacement; the exception Gallup flags is organisations with 10,000+ employees, where adopters report slightly more reductions (33%) than expansions (30%).

9. Why AI adoption numbers disagree: what not to compare

The commonest error in AI-at-work coverage is putting a worker survey next to a firm survey, or a "used it once this year" number next to a "most of my work" number, and calling the difference a trend.

The 40.7% Real-Time Population Survey and 78% Survey of Business Uncertainty figures are as reported in the Federal Reserve Board FEDS Note "Monitoring AI Adoption in the U.S. Economy" (April 3, 2026), which also shows the same Atlanta Fed survey at 69.4% when firms are weighted equally.

Two further cautions. First, the Census Bureau changed its question in November 2025 from AI use "in producing goods or services" to use "in any business function", and the Federal Reserve note shows sizeable level jumps across that break; never chart the old and new series as one line. Second, Gallup's organisational-integration series added a "don't know" option in Q3 2025, which is why the 44% integration figure from mid-2025 is higher than the 37% that followed.

10. What this means for HR and people teams

The frontline gap is the one most organisations will not see in their own dashboards, because the people on the wrong side of it are the least likely to answer a digital survey. If AI is meant to raise productivity across the business rather than inside head office, the 2.7× daily-use gap is the number to close. For the HR-specific side of AI — automation of reward and recognition workflows — see our guide to AI in HR and employee rewards automation, and for the wider HR tooling picture, our HR technology statistics. Training is the other half of the equation: our training and development statistics track what employers spend on it.

Methodology and sources

Every figure on this page was checked against the primary publication on 26 October 2026. Gallup figures come from the Gallup Panel, a probability-based panel of US adults working full or part time, weighted to Current Population Survey benchmarks; quarterly samples range from 18,871 to 23,717 with a margin of error of about ±1 point. Pew figures come from the American Trends Panel. Census figures come from the biweekly Business Trends and Outlook Survey of US employer businesses. We did not use vendor surveys whose sample or question wording we could not verify.

Derived metrics (ours, recomputable from the tables above): daily-user share of AI users (daily ÷ total, Gallup); remote-capable versus non-remote intensity multipliers (Gallup); the integration-to-plan ratio (47% ÷ 25%, Gallup May 2026); the worker-versus-firm gap (Gallup 52% ÷ Census 19.8%); and annual hours saved (St. Louis Fed share × 2,080 hours). Because inputs are published as rounded percentages, derived ratios carry rounding error of a few percent.

Key takeaways

• Half of US employees now use AI at work, 30% weekly and 15% daily — and the daily share is where the growth is.
• Total adoption has slowed to a point or two a quarter, while daily users rose from 20% to 28.8% of AI users in a year.
• The divide is widening by intensity: remote-capable workers are 2.06× as likely to use AI at all and 2.71× as likely to use it daily.
• Deployment is outrunning direction: 47% integrated, 25% with a clear plan, 36% with a strongly supportive manager.
• Only about one in five US businesses uses AI, but 37% of large firms do, which is why worker and firm figures differ 2.6×.
• Users save about 5.4% of their hours, and 31% say the saving has come back as more responsibility.