Anthropic Maps Three AI Economies, from Mild Boost to Mass Job Loss
Anthropic has modeled how AI could reshape the US economy by 2030, presenting three distinct scenarios: modest, substantial, and extreme.
Modest Scenario
In this scenario, AI’s impact is comparable to that of the internet. US GDP in 2030 is projected to be $34.1 trillion, a 1.6% increase from a no-AI path.
Substantial Scenario
Here, AI performs half of all knowledge work by 2030, with most tasks automated but not necessarily replaced. The economy grows at twice its normal rate to $36.3 trillion. While knowledge worker wages remain roughly stable, other workers experience gains, aligning with findings that AI primarily affects paychecks, not payrolls.
Extreme Scenario
This is the most dramatic case, where AI surpasses human productivity in most knowledge-work tasks and performs them autonomously. Annual GDP growth reaches a staggering 15%, doubling the economy every four and a half years, resulting in a 2030 GDP of $44.4 trillion. However, this scenario predicts significant job loss, with knowledge worker unemployment rising to nearly 18%, and overall unemployment surpassing typical recessionary levels.
Key findings:
- Unemployment: Knowledge workers face significant displacement, while other workers may have to transition to less AI-exposed roles.
- Wages: In the extreme scenario, knowledge worker wages fall by more than 10%, while wages for other workers increase by about a third.
- Income Distribution: The share of income going to workers decreases, while the share going to capital increases.
Anthropic’s model is interactive and explores these scenarios based on tasks performed by AI, its capabilities, and adoption rates.
Source:
Anthropic (2026). Economic Scenarios for Transformative AI (Korinek et al.)