日本語 ← Back to home
News

Microsoft Estimates Global AI Usage at 18.8 Percent

Microsoft Estimates Global AI Usage at 18.8 Percent. A look at the official announcement, practical implications and limits.

Article ID: TC-0020 Published:

Microsoft Estimates Global AI Usage at 18.8 Percent. A look at the official announcement, practical implications and limits.

The official announcement describes a development that may affect how people and organizations use digital tools. Details of rollout and availability should be checked against the original source.

WHAT 18.8% ACTUALLY MEANS: Microsoft's Global AI Diffusion Report estimates that 18.8% of people aged 15 to 64 worldwide used a generative AI product during the measurement period as of June 2026. This is not a share of the entire world population and does not mean 18.8% use AI every day.

QUARTERLY GROWTH: Microsoft reports an increase of roughly one percentage point from the first quarter of 2026, with growth in almost every economy. That does not establish a permanent growth rate. New products, pricing and access conditions can change adoption patterns.

THE LEADING COUNTRIES: The United Arab Emirates leads the reported figures at 73.3%, followed by Singapore at 64.3%. These are estimates for the working-age population under Microsoft's definition. High adoption alone does not prove higher productivity or better outcomes.

SOUTH KOREA AND SAUDI ARABIA: South Korea recorded the largest absolute increase, at 3.5 percentage points. Saudi Arabia moved five places from 30th to 25th in the ranking. Percentage-point changes and relative percentage increases describe different quantities, and rankings depend on other countries' performance.

JAPAN'S INCREASE: Microsoft says Japan rose by 2.2 percentage points from its first-quarter level of 22.5%, implying approximately 24.7% in the second quarter. That represents roughly 10% relative growth. The estimate does not establish which services people used or whether their work improved.

THE NORTH-SOUTH DIVIDE: Estimated usage reached 28.8% in the Global North and 16.2% in the Global South, a gap of 12.6 percentage points. Microsoft says the divide widened even as adoption rose. Global growth does not automatically produce equal access.

WHY ACCESS DIFFERS: Connectivity, devices, electricity, affordability, language support and digital skills all affect AI adoption. A product can be available in a country without being equally practical for everyone. Infrastructure and education are important parts of the story.

HOW MICROSOFT MEASURES DIFFUSION: The estimate comes from aggregated, anonymized Microsoft telemetry, adjusted for operating-system and device market shares, internet penetration and population. It is not a direct survey of every person. Like other modeled indicators, it depends on methodological assumptions.

A COMING MEASUREMENT CHANGE: Microsoft plans to include additional AI tools and models in its next report. It expects the broader measure to raise estimated usage in many economies. Future increases will therefore need to be separated into real behavioral change and changes to what the metric counts.

DIFFERENT USE CASES: Analysis of consumer Copilot activity suggests a relatively larger share of education and learning use in the Global South. The same adoption percentage can represent different activities across regions. Copilot usage should not be assumed to represent every AI product.

OPEN-WEIGHT MODELS: Microsoft notes that open-weight models are gaining share in API token volume. They may reduce access costs and support local-language adaptation. Compute resources, connectivity and technical skills remain barriers to making those benefits widely available.

THE BIGGER QUESTION: The 18.8% estimate indicates that generative AI use is expanding, but adoption alone is not a measure of accuracy, productivity or social benefit. Future assessments should combine usage rates with evidence about skills, applications, reliability and economic outcomes.

The practical impact depends on the use case. Organizations should assess data quality, permissions, costs and the ability to verify outputs rather than treating a product announcement as proof of results.

Read the linked primary source for the full announcement and any subsequent availability updates.

Source

Official company announcement ↗