· Microsoft AI Economy Institute · Report Translation · 4 min read
Global AI Adoption in 2025 — A Widening Digital Divide
Microsoft AI Economy Institute report: Singapore's AI adoption rate is 60.9%, second in the world. About one in six people globally now use generative AI, but the North-South gap is widening.
Original: Global AI adoption in 2025 — A widening digital divide Full report: Global AI Adoption Report (PDF) Published: January 8, 2026 English version: original Microsoft blog text.
In the second half of 2025, global adoption of artificial intelligence continued to climb, rising 1.2 percentage points over the first half. About one in six people worldwide now use generative AI tools — meaningful progress for a technology that has only recently entered mainstream use.
To track this trend, we define the “AI diffusion rate” as the share of the global population that has used a generative AI product within a given measurement window. The metric is based on aggregated, anonymized telemetry from Microsoft, adjusted for operating-system and device market share, internet penetration, and country populations. The methodology is described in detail in our AI diffusion technical paper.[1]
No single metric is perfect, and this one is no exception. Through the Microsoft AI Economy Institute, we continue to improve how we measure global AI diffusion, including how adoption differs across countries in priority areas like scientific discovery and productivity gains. This report relies on the strongest cross-country indicators currently available; we will add to it as new measures emerge and mature.
A Widening North-South Gap
Despite progress in AI adoption, the data show the gap is widening: adoption in the Global North is growing nearly twice as fast as in the Global South. Today, 24.7% of the working-age population in the Global North uses these tools, compared to just 14.1% in the Global South.
Leading Countries
Countries that invested early in digital infrastructure, AI skills training, and government adoption continue to lead — including the United Arab Emirates, Singapore, Norway, Ireland, France, and Spain.
The UAE has consolidated its position as global No. 1, with 64.0% of its working-age population using AI by the end of 2025, up from 59.4% in the first half. The UAE’s lead has now widened to more than three percentage points.
Singapore remains second globally, with an adoption rate of 60.9%.
Global AI adoption rankings (second half of 2025):
| Rank | Country / Region | Adoption rate |
|---|---|---|
| 1 | UAE | 64.0% |
| 2 | Singapore | 60.9% |
| 3 | Norway | 46.4% |
| 4 | Ireland | 44.6% |
| 5 | France | 44.0% |
| 24 | United States | 28.3% |
The American Paradox
The United States illustrates that leadership in innovation and infrastructure, while important, is not by itself sufficient to drive broad AI adoption. The US leads the world in both AI infrastructure and frontier model development, yet its working-age AI usage rank slipped from 23rd to 24th, with usage of just 28.3% — well behind smaller, more digitalized and AI-focused economies.
South Korea’s Rise
South Korea is the standout success story of the year-end. It surged 7 places in the global rankings, climbing from 25th to 18th, driven primarily by government policy support, improved Korean-language frontier model capabilities, and product features tightly tuned to local consumers. Generative AI is now widely used in schools, workplaces, and public services, and South Korea has become one of ChatGPT’s fastest-growing markets, prompting OpenAI to open an office in Seoul.[2]
The Rise of DeepSeek
A parallel development reshaping the global AI landscape in 2025 was the rapid rise of DeepSeek — an open-source AI platform that has gained substantial traction in markets long underserved by traditional providers. By releasing its models under an MIT open-source license and offering a fully free chatbot, DeepSeek has removed both the economic and technical barriers limiting access to advanced AI.
Unsurprisingly, its strongest adoption has been in China, Russia, Iran, Cuba, and Belarus. More notable, however, is DeepSeek’s rapid uptake in Africa, helped by strategic outreach and partnerships with companies such as Huawei.[3]
This rapid evolution highlights a dimension of the US-China AI competition that is becoming increasingly important — the race to drive adoption of each country’s own models. DeepSeek’s success reflects China’s growing momentum in Africa, a trend that is likely to accelerate further in 2026. The rise of DeepSeek also points to a broader truth: the global diffusion of AI is profoundly shaped by accessibility, and the next wave of users may well come from communities historically underserved by technological progress. The challenge ahead is to ensure that innovation spreads in ways that narrow rather than deepen the divide.
References
[1] A. Misra, J. Wang, S. McCullers, K. White, and J. L. Ferres, “Measuring AI Diffusion: A Population-Normalized Metric for Tracking Global AI Usage,” Nov. 2025, arXiv:2511.02781.
[2] “OpenAI Korea set to launch next month,” The Korea Times.
[3] S. Rai, L. Prinsloo, and H. Nyambura, “China’s DeepSeek Is Beating Out OpenAI and Google in Africa,” Bloomberg News.