AI Adoption Is Surging, but Facilities and Language Gaps Persist
- By John K. Waters
- 02/20/26
Expert system may be spreading out faster than previous waves of customer tech, but a report from Microsoft’s AI Economy Institute recommends its benefits are concentrating in a reasonably little set of countries, with infrastructure and language emerging as significant dividing lines.
The report, “AI Diffusion Report: Where AI is most utilized, established, and developed,”estimates that more than 1.2 billion people have utilized AI tools in less than 3 years– a rate it compares with earlier general-purpose technologies. It also argues that quick headline growth masks fundamental restrictions: “With more than 1.2 billion users in under 36 months, AI has ended up being the fastest-adopted innovation in human history.”
Measured by the share of working-age adults utilizing AI tools, the United Arab Emirates ranked first at 59.4%, followed by Singapore at 58.6%, Norway at 45.3%, and Ireland at 41.7%. The United States was noted at 26.3%, while China was noted at 15.4%.
The institute states its diffusion price quote relies in part on Microsoft’s view into software usage. “By analyzing aggregated and anonymized telemetry from over 1 billion Windows gadgets, we can approximate the frequency of AI-related activity across regions,” the report declared, adding that it adjusts for the fact that the dataset excludes non-Windows gadgets.
An associated Microsoft Research technical report (“Measuring AI Diffusion: A Population-Normalized Metric for Tracking Global AI Usage”) describes a population-normalized use metric built from telemetry and adjusted for gadget access and mobile scaling across 147 economies.
Even with connection rising worldwide, the report’s findings fit a wider pattern of digital spaces. The International Telecommunication Union approximates 5.5 billion people were online in 2024, but says about one-third of the world remains offline, with the hardest-to-connect populations focused in lower-income and rural areas.
Power gain access to is another restriction the report highlights as fundamental for data centers and day-to-day AI use. The World Bank’s electricity gain access to indicator puts international access above 90% recently, however with much lower protection in low-income economies and parts of Sub-Saharan Africa. The Microsoft report frames the consequence in geographic terms, saying, “AI adoption in the Worldwide North is roughly 23%, compared with only 13% in the Global South.”
On the supply side, the report argues that the calculate needed to construct and run sophisticated AI remains focused. It says, “Datacenter capacity stays greatly focused, with the United States and China accounting for approximately 86% of international compute,” and provides a markdown pointing out International Energy Firm (IEA) price quotes, listing 53.7 gigawatts for the United States and 31.9 gigawatts for China.
The IEA has actually also warned that AI is set to greatly increase electrical power need from data centers over the coming years, intensifying pressure on grids in the biggest data center areas.
The report likewise attempts to different countries that build frontier designs. It says just 7 nations host “frontier-level” AI designs and that the efficiency gap is narrowing. In a table comparing each country’s finest design versus the frontier, it notes the United States at 0 months to frontier, China at 5.3 months, South Korea at 5.9 months, France at 7.0 months, the United Kingdom at 7.7 months, Canada at 7.8 months, and Israel at 11.6 months.
Among the report’s more pointed claims is that language can be a standalone barrier, even after accounting for earnings and connectivity. “Countries where low-resource languages are primary exhibit significantly lower AI adoption, even after managing for GDP and internet gain access to,” the report asserts.
That argument provides proof that widely used web corpora used in AI development are greatly skewed toward a little set of languages, with Typical Crawl’s language data showing a big concentration in leading languages such as English.
The institute’s bottom line is that diffusion is not simply a question. Ultimately, the value of expert system will be evaluated not by the variety of models produced, but by the level to which they benefit society,” the report concluded.
The complete report is available here on the Microsoft website.
About the Author
John K. Waters is the editorial director of a number of Converge360.com websites, with a focus on high-end development, AI and future tech. He’s been blogging about advanced innovations and culture of Silicon Valley for more than 20 years, and he’s composed more than a lots books. He likewise co-scripted the documentary film Silicon Valley: A 100 Year Renaissance, which aired on PBS. He can be reached at [e-mail protected]