Measuring AI Diffusion: A Population-Normalized Metric for Tracking Global AI Usage

TL;DR

This paper tackles the lack of population-normalized country-level AI usage data by introducing AI User Share, a real-time diffusion metric derived from Microsoft telemetry and scaled to the working-age population across 147 economies. The method combines the share of AI-using Microsoft users, desktop device penetration, and mobile usage scaling, with adjustments for opt-in bias and data sparsity. Key findings show a strong positive relationship between GDP per capita and AI adoption, substantial latent demand among internet-connected populations in lower-income countries, and rapid, event-driven shifts in adoption driven by product launches such as DeepSeek in 2025. The metric offers policymakers a timely, cross-country benchmark to monitor AI diffusion, identify infrastructure gaps, and guide inclusive AI policy, while noting biases inherent to telemetry-based data sources.

Abstract

Measuring global AI diffusion remains challenging due to a lack of population-normalized, cross-country usage data. We introduce AI User Share, a novel indicator that estimates the share of each country's working-age population actively using AI tools. Built from anonymized Microsoft telemetry and adjusted for device access and mobile scaling, this metric spans 147 economies and provides consistent, real-time insight into global AI diffusion. We find wide variation in adoption, with a strong correlation between AI User Share and GDP. High uptake is concentrated in developed economies, though usage among internet-connected populations in lower-income countries reveals substantial latent demand. We also detect sharp increases in usage following major product launches, such as DeepSeek in early 2025. While the metric's reliance solely on Microsoft telemetry introduces potential biases related to this user base, it offers an important new lens into how AI is spreading globally. AI User Share enables timely benchmarking that can inform data-driven AI policy.

Figures (6)

• Figure 1: Global distribution of AI adoption rates by economy. The United Arab Emirates and Singapore exhibit the highest adoption, with over half of the working-age population using AI tools. Most high-adoption economies are in Europe and North America.
• Figure 2: Regional comparison of AI user share as of June 2025. North America and Europe and Central Asia exhibit the highest adoption rates, while South Asia and Sub-Saharan Africa remain below 13%.
• Figure 3: Technology adoption vs GDP per Capita for electricity, internet access, digital skills and AI User Share. All four technologies are highly correlated with GDP per Capita, though the absolute level of adoption decreases as we progress from electricity access to AI User Share.
• Figure 4: Comparison of AI Adoption Rates with GDP Per Capita. In general, there is an upward trend with higher AI adoption as GDP per capita increases. However, there appears to be a leveling off in the range of 25-40%, implying a potential ceiling to AI adoption given current conditions.
• Figure 5: AI User Share as a percent of the connected population for countries with the lowest internet penetration. When normalized to population connected to the internet, many of these countries show high rates of AI adoption.