AI adoption is growing, but the latest figures do not describe one universal adoption rate. U.S. surveys count firms, workers, tasks or business functions; global estimates measure population use; market totals and energy figures may be forecasts rather than final measurements. The data below keeps those distinctions visible so the numbers are useful without pretending they are directly comparable.
How widely are U.S. businesses and workers using AI?
The answer depends on who is counted and what qualifies as AI use. The Federal Reserve’s April 2026 synthesis discusses several U.S. measures with different denominators; Census Bureau data adds detail about business functions and work tasks.
As an Amazon Associate I earn from qualifying purchases.
| Measure | Finding | What it counts |
|---|---|---|
| Business AI use at year-end 2025 | 18% of firms | U.S. Census Bureau Business Trends and Outlook Survey (BTOS) estimate, as reported by the Federal Reserve in April 2026. |
| AI adoption among employers | 78% employment-weighted; 54% for large language models | November 2025 Atlanta Fed Survey of Business Uncertainty estimates. These are shares of workers employed at firms reporting adoption, not shares of firms. |
| Individual work-related generative AI use | About 41% | Workers reporting work-related generative AI use in the November 2025 Real-Time Population Survey. |
| AI in at least one business function | 18% of firms; 32% employment-weighted | U.S. Census Bureau AI supplement, November 2025–January 2026 reference period. Firms expected the firm-weighted rate to reach 22% within six months. |
| Workers using AI for work-related tasks | 23% of firms; 41% employment-weighted | U.S. Census Bureau AI supplement. This is a firm survey, not the same measure as the individual-worker survey above. |
These rates should not be collapsed into a single adoption percentage. A firm-weighted measure gives a small business the same weight as a large one; an employment-weighted measure reflects how many workers are at adopting firms. Individual surveys instead ask people about their own use. The Federal Reserve notes that survey wording, respondent, weighting and definitions also contribute to differences.
Free tools Windows power users keep installed
One-click scans. No signup required.
Adoption is often narrow rather than company-wide
Among firms using AI in the Census supplement, 57% used it in three or fewer business functions. The most commonly reported functions were sales and marketing (52%), strategy and business development (45%), and IT (41%). These percentages describe functions among AI-using firms, not the share of all firms using AI in each function.
#1 Best Overall
For work-related generative AI tasks, writing, document analysis and information search were the leading uses. Sixty-five percent of firms reported limiting use to three or fewer tasks. The Census analysis found that 66% of users relied on AI solely to augment tasks; AI-related employment decreases were reported by 2% of firms. Those survey findings do not establish that AI cannot displace work.
How much are businesses and governments spending on AI?
Market totals vary by scope. Gartner’s September 16, 2026 figures are forecasts for worldwide spending, not final measured totals. Its narrower market for AI models and platforms is a separate category and should not be added to the overall total.
Rank #2
| Category | 2026 figure | How to read it |
|---|---|---|
| Total AI spending | $2.7 trillion; 49.5% year-over-year growth | Gartner worldwide forecast published September 16, 2026. |
| AI infrastructure | $1,484,397 million | Gartner’s 2026 forecast segment within its overall AI spending classification. |
| AI software | $461,637 million | Gartner’s 2026 forecast segment within its overall AI spending classification. |
| AI models and platforms | $64.252 billion; 63.4% growth from 2025 | Gartner worldwide end-user spending forecast published July 20, 2026; a narrower market than total AI spending. |
Model growth forecasts point to specialized products
In its July 2026 forecast, Gartner projected 117% growth in generative AI models and 210% growth in domain-specific and specialized generative AI models for 2026. These are forecast growth rates for market segments, not measured growth already achieved. Gartner’s model-and-platform total and the September total AI spending forecast cover different market definitions.
AI investment and GDP estimates answer different questions
The International Monetary Fund’s 2026 annual report page says AI-related technology investment added an estimated 0.5 percentage point to U.S. GDP growth in 2025. Separately, it cites external estimates that private-sector-driven global AI investment could exceed $2 trillion in 2026. The first is an estimate of a contribution to U.S. growth; the second is an externally attributed global investment estimate, not the IMF’s own spending forecast.
How does generative AI use vary around the world?
The Microsoft AI Economy Institute estimated that 17.8% of the global population used generative AI in the first quarter of 2026. Its regional estimates were 27.5% in the Global North and 15.4% in the Global South. These are Microsoft estimates based on its methodology, not a census of every AI tool or all AI use.
Compared with the second half of 2025, Microsoft reported increases of 2.8 percentage points in the Global North and 1.3 points in the Global South. The report connects the difference in diffusion with disparities in electricity access, internet connectivity and digital skills. That is the report’s interpretation of the gap, not proof that any one factor alone caused it.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.What do AI and data centers mean for electricity use?
The International Energy Agency reported that electricity consumption by data centers grew 17% in 2025, while consumption by AI-focused data centers grew 50%. Those figures show why efficiency gains per task do not automatically mean lower total demand: use can grow, and workloads differ substantially.
Separate observed growth from projected demand
In the IEA’s central projection, total data-center electricity consumption rises from 485 terawatt-hours in 2025 to 950 TWh in 2030, approximately 3% of projected global electricity demand in 2030. The 2030 figure is a projection, not observed consumption.
Best Value
The IEA also estimated that capital expenditure by five large technology companies exceeded $400 billion in 2025 and expected it to increase a further 75% in 2026. The 2026 increase is an estimate, not a final spending result.
Energy per query is not a fixed number
The IEA says energy use per individual AI task has fallen by at least an order of magnitude annually in recent years. At the same time, video generation, reasoning and agentic workloads can consume hundreds or thousands of times more energy per query than simple text generation. A single average “energy per AI query” would therefore conceal major differences between tasks.
Do the figures show that AI is improving business performance?
They show adoption and investment, but they do not yet establish a general causal link between AI and improved business outcomes. The U.S. Bureau of Economic Analysis’s 2026 paper describes a changing pattern in business expectations: adoption initially grew more slowly than expected, then faster than expected for a short period, and more recently tracked expectations more closely. It finds some evidence connecting adoption motivations with production-process changes and increased research and development intensity, while saying the connection to observed outcomes remains unclear.
Quick wins for a faster PC:
Repair Windows errors before they cause bigger problemsFix Now →Fix the driver behind crashes, sound loss and screen glitchesFind Drivers →Clear out junk files and repair common Windows errorsFree Scan →The Census Bureau found a positive correlation between the breadth of AI integration and commercial performance. Correlation does not establish that AI caused better performance; firms with stronger performance may differ in other ways, too. The analysis also found that relationships with employment decreases varied by type of integration.
How should you interpret AI statistics for 2026?
- Check the unit. A percentage of firms, workers, employment, tasks or population is not interchangeable with another.
- Check the scope. “AI,” generative AI, large language models and AI in a particular business function describe different things.
- Check the period and geography. A U.S. survey reference period is not a global estimate, and publication date may come after the period measured.
- Check whether a number is observed or forecast. Gartner’s 2026 spending totals and the IEA’s 2030 electricity figure are projections, while the 2025 electricity growth rates are reported observations.
- Check the claim being made. Adoption does not by itself prove productivity gains, job losses or improved financial performance.
The IEA says comprehensive worldwide statistics on the frequency and depth of AI use are unavailable. The figures here are consequently best read as distinct snapshots from named sources, rather than pieces of one complete global scoreboard.
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.
