Microsoft published its latest Global AI Diffusion Report on May 7, 2026, painting a picture of generative AI moving from early-adopter curiosity to mainstream working tool — but unevenly. The report finds that 17.8% of the world's working-age population used a generative AI product in the first quarter of 2026, up from 16.3% the prior period, a 1.5-percentage-point bump that translates into hundreds of millions of additional users over a single quarter.
The headline of this edition is breadth. Twenty-six national economies now exceed 30% AI usage among their working-age population, a threshold that signals AI has crossed from novelty into routine workflow. At the top of Microsoft's National AI Leaderboard sits the United Arab Emirates at 70.1%, a remarkable concentration that reflects state-led adoption pushes and a heavily digital workforce.
The U.S. inches up the leaderboard
The United States moved from 24th to 21st place, with 31.3% of its working-age population now using generative AI products, according to Microsoft's telemetry-based methodology. That ranking — well behind Gulf states, several European economies, and parts of East Asia — has become a recurring point of awkwardness for U.S. policymakers given that most frontier model labs are headquartered domestically.
The report attributes the strongest quarterly gains in Asia to improved multilingual capabilities, with South Korea, Thailand, and Japan posting the biggest movement on the leaderboard. As frontier models close the gap on languages other than English, adoption is following.
A widening Global North / Global South gap
Microsoft's data shows the Global North averaging 27.5% adoption versus 15.4% in the Global South — a gap that has widened, not narrowed, even as overall numbers rise. The company frames this as the central diffusion challenge of the next several years: tools are getting more capable, but access is still gated by infrastructure, language coverage, and device economics.
Methodologically, the report defines diffusion as the share of people aged 15–64 who used a generative AI product in the reported period, derived from Microsoft telemetry adjusted for operating system share, device penetration, internet access, and population.
Developers stay employed — and busier
One of the more closely watched sections covers software developers, the cohort whose jobs frontier AI was supposed to absorb first. Microsoft reports that global git pushes are up 78% year over year, and that U.S. software developer employment hit roughly 2.2 million in 2025, an 8.5% annual increase that marks a record for the profession. Early 2026 data shows March developer employment running about 4% above the prior March.
That cuts against the narrative — pushed by some lab CEOs and amplified by recent layoff cycles — that AI is rapidly displacing software engineers. Microsoft's data suggests that, at least in aggregate, AI is amplifying developer output rather than shrinking the headcount needed to ship code.
Why it matters
The Diffusion Report has become one of the few quarterly datasets policymakers, investors, and enterprise buyers cite when arguing about how fast AI is actually landing in workplaces. The May 7 edition provides ammunition to both camps: bulls will point to the 1.5-point quarterly jump and 26 countries past 30%, while skeptics will note that more than 80% of working-age adults globally still don't use generative AI on any given week. Both readings are correct — and both will shape the next round of capex, regulation, and product roadmaps.
