Global AI adoption inequality hardening as high-income countries consolidate usage share while US regional convergence slows

str 5 12/31/2099 · 1 article

structural · economic · AI · US, Global

Analysis

International AI usage concentration increased (Gini 0.46→0.50, top 20 countries 45%→48% of per-capita usage) while US state-level convergence decelerated from 2-5 year equalization timeline to 5-9 years. This divergence suggests global adoption barriers are strengthening even as domestic US inequality narrows.

Key actors

Anthropichigh-income countries

Source article

Economic-Index-5-v2 (1)

pdf-archive — Work, Skills and Society · — · extracted in run pdf-import-undated-1779256502138-107 · 5/20/2026, 7:17:00 AM

"the top 20 countries account for 48% of all per-capita usage, up from 45%, underscoring a persistent gap in global adoption" [48% of all per-capita usage, up from 45%]

The quantified increase in concentration of per-capita usage among top 20 countries directly measures the hardening of global inequality in AI access, contradicting the within-US convergence trend.

Reasoning from this article

The report shows a bifurcated adoption pattern: within the US, lower-usage states are catching up (Gini falling, top-5 share 30%→24%), but globally, high-income countries are pulling further ahead. This suggests that domestic policy (CHIPS Act, regional incentives) may be equalizing US access while international barriers (cost, infrastructure, regulatory fragmentation) are widening global gaps. The slowdown in US convergence pace (from 2-5 to 5-9 years) indicates the easy gains are exhausted and remaining inequality is more structural.