Not according to the best available evidence. A late-2025 Census survey result showed a small decline in the share of businesses reporting very recent AI use, but it did not measure the number of individual workers using AI. The Census question also changed in November 2025. Newer Census data through May 2026 show adoption rising among larger firms, while Gallup found that more employees reported using AI at work. The defensible conclusion is that workplace AI adoption is uneven and selective—not suddenly collapsing.
What the headline gets wrong
The headline turns a narrow business-survey movement into a claim about people. Those are different measurements.
| Metric | What it measures | Why it matters |
|---|---|---|
| Businesses reporting any AI use | Whether a firm used AI during the survey period | Shows firm adoption, not how many employees participate |
| AI in production or service delivery | Use in the older Census question’s narrower scope | Can miss administrative and support functions |
| Employees personally using AI | Workers’ self-reported use | Captures informal or unauthorized use |
| Employer-approved tool use | Use of sanctioned systems | May exclude consumer tools used privately |
| Task-level use | Whether workers apply AI to particular duties | Shows where adoption is actually embedded |
A large employer counts once in a firm survey but may employ thousands of people. Conversely, employees can use ChatGPT, Claude or Gemini for drafting, research or coding without the company reporting AI in production. Census research found this gap directly: 18% of firms reported AI use in a business function during its November 2025–January 2026 supplement, versus 32% when results were weighted by employment; 23% said workers used AI in work-related tasks. See the Census working paper.
What the late-2025 decline actually showed
Futurism reported that approximately 11% of large companies said they had used AI to “produce goods and services” in October 2025, down from about 12% in the preceding survey period. It also reported that the share of businesses with 100–249 employees saying they had not used AI in the previous two weeks rose from 74.1% in March to 81.4% in the cited result. For businesses with more than 250 employees, the comparable no-use figure reached 68.6%, compared with a 2025 low of 62.4% in February. These are historical observations from the older series, not a current count of people. The original coverage is at Futurism.
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 →Clear out junk files and repair common Windows errorsFree Scan →#1 Best Overall
A one-percentage-point change between adjacent survey periods can reflect sampling variation, seasonality or a temporary pause. It is not enough by itself to establish a structural reversal.
The Census question changed
The Business Trends and Outlook Survey (BTOS) began publishing core AI-use questions in October 2023. Until the collection period beginning November 17, 2025, its main question asked whether a business had used AI “in producing goods or services.” The revised question asks whether AI was used “in any of its business functions,” including finance, human resources, customer service, marketing, information technology and research and development.
Census says the wording change created a level shift and requires a new time series. The old and new series therefore should not be drawn as one uninterrupted trend line. The agency’s explanation is in its AI question wording update; BTOS methodology is described at the Census BTOS page.
- Old scope: AI used in producing goods or services.
- New scope: AI used in any business function.
- Change date: Collection period beginning November 17, 2025.
- Interpretation: A movement across the break may reflect wording as well as behavior.
What newer Census data show
Census analysis of data collected from December 14, 2025, through May 3, 2026 found that AI use increased among firms with at least 20 employees. Overall business use stayed roughly between 17% and 20% during that period. In the period ending May 3, 2026, 37% of firms with at least 250 employees and 32% of firms with 100–249 employees reported AI use in business operations. The full analysis is available in Census’s May 2026 report.
Do these 3 things before closing this tab:
1Scan for outdated or missing drivers - takes under a minute2Clear out junk files and repair common Windows errors3Fix the driver behind crashes, sound loss and screen glitches| Business group | Reported result | Reference |
|---|---|---|
| All covered businesses in the newer series | Approximately 17%–20% overall use from December 2025 to May 2026 | Census May 2026 analysis |
| 250 or more employees | 37% using AI in business operations, period ending May 3, 2026 | Census May 2026 analysis |
| 100–249 employees | 32% using AI in business operations, period ending May 3, 2026 | Census May 2026 analysis |
These figures do not prove that AI is delivering productivity gains. They do show that “companies are abandoning AI” is not supported by the latest official trend.
Employee surveys point upward
Gallup measures workers rather than businesses. Between the second and third quarters of 2025, the share of U.S. employees saying they used AI at work at least a few times per year rose from 40% to 45%. Frequent use increased from 19% to 23%, and daily use from 8% to 10%, according to Gallup’s December 2025 results.
Gallup also found a sharp occupational split. In June 2025, 27% of white-collar employees reported frequent AI use, compared with 9% of production and frontline workers. By December, at least-a-few-times-per-year use reached 76% among technology or information-systems employees, 58% in finance and 57% in professional services, with lower rates in retail, healthcare and manufacturing. The occupational comparison appears in Gallup’s June 2025 report.
Why adoption can look lower without a collapse
A two-week window misses intermittent use
BTOS asks about the previous two weeks. A business can maintain an AI program, use it monthly and still answer “no” during a quiet fortnight.
Free tools Windows power users keep installed
One-click scans. No signup required.
Experiments may be replaced by narrower deployments
After initial pilots, firms may stop novelty testing, consolidate vendors or restrict AI to tasks with a measurable return. Census’s 2026 microstructure study found that 65% of firms limited AI use to three or fewer tasks. That pattern is consistent with filtering, although it does not prove that “AI fatigue” caused any decline.
Embedded features hide the tool switch
An employee may stop opening a stand-alone chatbot after similar capabilities appear inside Microsoft 365, Google Workspace, a CRM, an integrated search tool or a coding environment. Product usage can fall while underlying AI-assisted work continues.
Informal use is hard to count
Workers may use personal accounts without employer approval. Procurement data will miss that activity, while employee surveys may capture it. The reverse problem also occurs: spam filtering, fraud detection, optical character recognition, search ranking and predictive maintenance may use AI without employees labeling the software “AI.”
Firm size changes the denominator
Firm-level adoption and employment-weighted adoption answer different questions. A small business and a 10,000-person employer each count as one firm, which is why employment-weighted use can be substantially higher.
How to read any workplace-AI statistic
- Population: Is the sample U.S. businesses, U.S. employees or global workers? Which industries and occupations are included?
- Unit: Are results based on firms, people, tasks, purchased seats, active users, messages or outputs?
- Time period: Does “use” mean once a year, in the last two weeks, weekly or daily?
- Definition: Does AI include generative tools, predictive analytics, automation and embedded software?
- Survey design: Were questions, weighting, sample or response methods changed? Is the difference statistically meaningful?
- Business meaning: Is the activity a pilot, production workflow, employer-approved use or unsanctioned experimentation?
What this means for employees, investors and buyers
Employees should not infer from the late-2025 headline that AI skills are becoming irrelevant. Use is growing fastest in technology, finance, professional services and other document-heavy roles, while physical and frontline work is changing more slowly.
Business leaders should measure recurring task use rather than licenses. Before expanding a program, track weekly active users, completed tasks, time saved, quality, error rates and the share of use concentrated among a few specialists. Confirm that data permissions, retention and compliance controls match the work.
For buyers comparing a stand-alone assistant with Microsoft 365 Copilot, Google Workspace with Gemini, ChatGPT Business or Enterprise, or Claude Team or Enterprise, workflow fit matters more than a generic “most powerful” label. The official entry points are OpenAI Business, Microsoft 365 Copilot, Google Workspace AI and Anthropic Enterprise. Pricing and plan limits can change, so verify them directly before purchase.
A practical test for your own organization
- Count employees using AI weekly, not just licensed seats.
- Separate employer-approved tools from personal accounts.
- List the specific tasks supported and how often each recurs.
- Record baseline time, quality and error rates before a pilot.
- Measure completed work and outcomes, not merely prompts or logins.
- Check whether a tool change moved AI into an existing workflow rather than eliminating use.
- Consolidate products that do not generate recurring, measurable value.
The Bottom Line
The evidence supports a more precise verdict: workplace AI adoption is uneven and may be cooling in some experiments, but there is no reliable evidence that the overall number of people using AI at work suddenly fell. The late-2025 headline overread a short, firm-level measure whose question changed, while newer business and employee data point to continued—though selective—growth.
Recommended Free Tools
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.
