Global AI Usage Hits 18.8% as North-South Gap Widens

Microsoft has released its latest Global AI Diffusion Report, indicating that artificial intelligence adoption is continuing to expand across the globe. As of June 2026, approximately 18.8% of the working-age population worldwide had used a generative AI product, marking a modest increase of about one percentage point since the start of the year. The growth has been broad-based, with nearly every economy reporting higher usage rates compared to the first quarter.

Despite this general upward trend, the data highlights a deepening divide between the Global North and the Global South. Usage in the North has reached 28.8%, while it stands at 16.2% in the South. This disparity suggests that while AI is becoming a global utility, the benefits and accessibility of the technology remain unevenly distributed, leaving a significant portion of the world's population with less exposure to these tools.

Regional leaders drive adoption growth

The United Arab Emirates and Singapore continue to lead the rankings, with 73.3% and 64.3% of their respective working-age populations using AI. South Korea recorded the largest absolute increase in usage, gaining 3.5 percentage points over the quarter. Saudi Arabia also made a notable leap, moving up five places in the global rankings to reach 25th, while Japan posted the largest relative growth rate at roughly 10%.

These figures indicate that adoption is not uniform. Some nations are rapidly closing the gap, while others remain at lower levels. The report notes that Japan’s growth, though significant in relative terms, starts from a lower base than the leaders, highlighting the different stages of AI maturity across different economies.

Measurement methods face limitations

The data is derived from anonymized Microsoft telemetry, adjusted for market share and internet penetration. However, The Official Microsoft Blog acknowledges that no single metric is perfect. The current measure only captures generative AI products and does not yet include newer tools and models that have emerged in the past year. Microsoft expects to expand its methodology in the next report, which will likely show higher usage figures, particularly in China.

This limitation means the current figures may understate the true extent of AI integration. By excluding non-generative tools, the report provides a snapshot of a specific slice of the AI ecosystem rather than a comprehensive view of all AI interactions. Users should view these percentages as a baseline for generative AI specifically, not the entire spectrum of artificial intelligence applications.

Usage patterns differ by region

Beyond simple adoption rates, the report reveals distinct differences in how AI is used in the Global North versus the South. In the Global South, there is a notably higher share of usage dedicated to education and learning. This suggests that in many developing economies, AI is serving as a critical tool for knowledge acquisition and skill development, whereas in the North, it may be more heavily integrated into productivity and professional workflows.

The rise of open-weight AI models also plays a role in this dynamic. These models are capturing a growing share of API usage, which could lower access costs and allow for adaptation to local languages. However, significant barriers such as infrastructure, connectivity, and technical skills remain, preventing these potential benefits from being fully realized in many regions.