# anthropic-economic-index-adoption-inegale-2025-09-15

## Veille

Uneven AI Adoption - Geographic and Enterprise Disparities - Anthropic Economic Index

## Titre Article

Rapport de l'Indice Économique d'Anthropic : Adoption inégale de l'IA au niveau géographique et en entreprise

## Date

2025-09-15

## URL

https://www.anthropic.com/research/anthropic-economic-index-september-2025-report

## Keywords

AI adoption, economic impact, geographic distribution, enterprise AI, Claude.ai, API usage, automation, augmentation, labor market, economic inequality, digital divide, O*NET tasks, Anthropic Economic Index, Indice d'Utilisation de l'IA d'Anthropic

## Authors

Ruth Appel, Peter McCrory, Alex Tamkin, Miles McCain, Tyler Neylon, Michael Stern

## Ton

**Profile:** Academic-professional | Research institutional | Analytical-educational | Expert

The report adopts a rigorous research-institutional tone, combining data science with public policy analysis. The multiplicity of authors reflects a collaborative research team. The specialized terminology (AUI index, O*NET tasks, directive vs. augmentative usage) targets an audience of researchers, policymakers, and business analysts. The systematic methodological structure reflects academic rigor. Precise quantitative citations (40% adoption, 77% automation) empirically anchor the claims. The framing around economic impact and policy implications signals a dual audience: the research community AND policymakers. The open-source release of the data reflects a commitment to reproducible research. This is a style typical of AI research reports combining scientific rigor with practical relevance.

## Pense-betes

- 40% of US employees use AI at work (vs. 20% in 2023)
- Increase in "directive" tasks (full delegation) on Claude.ai
- Automation surpasses augmentation for the first time
- Strong correlation between usage and income across countries
- Singapore and Canada lead in per capita usage
- Washington D.C. and Utah dominate in the United States (ahead of California)
- Low-adoption regions: focus on coding; high-adoption regions: diversified usage
- 77% of business usage (API) involves automation
- Businesses: specialized usage, predominantly automation
- Coding and office/administrative tasks predominant in enterprise
- Bottleneck: access to appropriate contextual information
- Risk of exacerbating existing economic inequalities
- Data released open source for independent research

## RésuméDe400mots

The Anthropic Economic Index report examines the rapid and uneven diffusion of AI, particularly Claude, across different geographic areas and business sectors. The study finds that AI adoption is occurring at an unprecedented pace, with 40% of US employees reporting using AI at work, up from 20% in 2023.

Analysis of conversations on Claude.ai reveals significant shifts in usage over time, with an increase in tasks related to education and science. Crucially, users are granting Claude greater autonomy in "directive" conversations, where tasks are fully delegated. This shift suggests an acceleration toward automation, with directive usage surpassing augmentation (collaborative usage) for the first time.

Geographically, AI adoption is concentrated in high-income, technologically advanced countries. The Indice d'Utilisation de l'IA d'Anthropic (AUI) shows a strong correlation between per capita Claude usage and income across countries. Singapore and Canada lead in usage relative to population, while emerging economies such as India and Nigeria show lower adoption.

In the United States, Washington D.C. and Utah dominate per capita usage, even surpassing California, traditionally regarded as the tech hub. Low-adoption regions focus more heavily on coding tasks, while high-adoption areas show more diversified applications in education, science, and business. Paradoxically, high-adoption countries show less automated and more augmented usage, suggesting a preference for collaborative human-AI interaction as adoption matures.

Regarding enterprise adoption, analysis of first-party API traffic reveals that businesses use Claude in a specialized way, predominantly for automation. Coding and office/administrative tasks predominate, and 77% of business usage involves automation patterns, a figure far higher than among individual Claude.ai users.

Model capabilities and the economic value of automation appear to drive adoption more than cost, as more expensive tasks often show higher usage rates. A key bottleneck identified for sophisticated enterprise deployment is access to appropriate contextual information. Complex tasks require long inputs, implying a need for data modernization and significant organizational investment.

The report concludes that uneven patterns of AI adoption risk exacerbating existing economic inequalities if productivity gains concentrate in already-prosperous regions and automation-ready sectors. The open-source release of the data aims to encourage independent research into the economic impacts of AI and to inform policy responses that ensure broader benefits. The authors emphasize that while technical capabilities are advancing rapidly, it is society's policy choices that will ultimately shape the future economic effects of AI.

## GrapheDeConnaissance

- Anthropic —publie→ Anthropic Economic Index (DOCUMENT, 0.98)
- Ruth Appel —a_créé→ Anthropic Economic Index (DOCUMENT, 0.97)
- Anthropic Economic Index —mesure→ adoption géographique et entreprise de l'IA (CONCEPT, 0.96)
- adoption IA au travail —mesure→ 40% des employés américains (MESURE, 0.97)
- usage directif —surpasse→ usage augmentatif (CONCEPT, 0.92)
- Singapour —mesure→ usage Claude par habitant parmi les plus élevés (MESURE, 0.93)
- Canada —mesure→ usage Claude par habitant parmi les plus élevés (MESURE, 0.91)
- Washington D.C. —surpasse→ Californie (LIEU, 0.9)
- entreprises clientes API —mesure→ 77% d'usages automatisés via API (MESURE, 0.95)
- Anthropic Economic Index —prédit→ l'adoption inégale de l'IA risque d'exacerber les inégalités économiques existantes (AFFIRMATION, 0.88)
- Anthropic —publie→ Anthropic Economic Index (DOCUMENT, 0.94)
- accès aux données contextuelles —est_instance_de→ goulot d'étranglement déploiement entreprise (CONCEPT, 0.87)
- Indice d'Utilisation de l'IA (AUI) —mesure→ corrélation usage-revenu entre pays (MESURE, 0.9)

---
Canonical: https://www.thekb.eu/en/fiches/anthropic-economic-index-adoption-inegale-2025-09-15/