America builds a great deal of the world’s AI, yet it does not top every table measuring who uses it. That is one of the interesting tensions in the 2026 Stanford AI Index. It is also a story that can go badly wrong if we mix up the tables.
Two numbers invite that mistake: roughly 53% adoption in a comparison with earlier technologies, and 28.3% for the United States in an international ranking. They are not a global average and an American result from the same survey. The statistics have arrived at the party wearing matching outfits, but they are not twins.
What the 28.3% Measures
The international figure comes from Microsoft’s January 2026 diffusion report. It estimates the share of the working-age population using generative AI in the second half of 2025, using aggregated, anonymized telemetry with adjustments for factors including device market share and internet penetration.
On that measure, the U.S. ranks 24th at 28.3%. The United Arab Emirates is at 64%, and Singapore at 60.9%. The same report puts the worldwide estimate at 16.3%, rather than 53%.
So the U.S. trails several countries in this ranking while remaining above the report’s global estimate. “America barely uses AI” is a much stronger claim than those figures support. Nor does this measure tell us that everyone outside the 28.3% has never tried a chatbot.
Faster Than the Internet—Under Which Comparison?
Stanford’s roughly 53% figure belongs to a different adoption comparison, drawing on survey-based research about the uptake of generative AI relative to PCs and the internet. It should not be relabeled as a census of the world’s population.
A useful companion is the St. Louis Fed’s November 2025 account of this research. Its U.S. survey covers adults aged 18–64 and reports 54.6% using generative AI for work or nonwork purposes in August 2025, following a revision to the survey methodology.
The authors compare that with estimated PC and internet adoption three years after their chosen mass-market starting points. Their result supports rapid early uptake in that comparison. It does not prove that every country is adopting AI at the same speed, or that AI has outrun every technology by every possible definition.
Surveys and telemetry answer related questions through different windows. The sensible response to different percentages is to inspect those windows, not quietly combine whichever numbers make the loudest headline.
Big Investment, Different Everyday Experience
Stanford reports $285.9 billion in U.S. private AI investment for 2025, versus $12.4 billion in China. Its economy chapter cautions that private investment alone understates China’s broader funding picture. These figures cannot tell us which country buys an equivalent unit of AI capability more efficiently.
The chapter also reports an estimated $172 billion in annual U.S. consumer surplus from generative AI by early 2026. Consumer surplus describes estimated value to users beyond what they pay; it is not revenue collected by AI companies, cash deposited in users’ accounts or a directly measured contribution to GDP.
That leaves an interesting story without the statistical gymnastics. A country can host major AI companies while adoption remains uneven. A free tool can provide value without generating equivalent subscription revenue. And someone can try AI once without making it an essential part of everyday life.
What Matters Beyond the Leaderboard
For ordinary users, access is only a beginning. A useful tool must fit a real task, give understandable results and earn enough trust to be used again. An international ranking cannot explain all those individual decisions.
If you want a more concrete starting point, our guide to running AI locally looks at an alternative to relying entirely on cloud services. Trying a small task is more informative than feeling obliged to join a national adoption race.
The takeaway is still striking: generative AI is spreading quickly, but building it, trying it and benefiting from it are different achievements. The cat, meanwhile, continues to rank first in adopting any technology that produces a warm surface.
