KOSUNI CRYPTO
Korea's Crypto Pulse, in English
$BTC $ETH $XRP $SOL $DOGE
Macro
Neutral

Anthropic Releases AI Economic Scenarios for 2030, Warns of Labor Disruption

Published September 23, 2026 2:55 AM · 8 views
Anthropic Releases AI Economic Scenarios for 2030, Warns of Labor Disruption

Anthropic has released a model predicting how artificial intelligence could reshape the US economy by 2030. Notably, in its fastest-growing scenario, knowledge worker wages could fall by more than 10%.

The company's economics team unveiled the tool alongside a report titled 'Economic Scenarios for Transformative AI,' allowing users to input their own forecasts.

The model presents three distinct scenarios. The release follows criticism from former Anthropic researcher Jacob Coxon, who resigned and accused the industry of focusing on self-improving superintelligence development.

In a moderate scenario, AI grows at a scale similar to the internet, lifting GDP by $341 billion (a 1.6% increase). In a realistic scenario, AI handles half of all knowledge work by 2030, pushing GDP to $363 billion (double typical growth rates), with unemployment around 5% and stable wages for knowledge workers.

In an extreme scenario where AI self-improves without human assistance, annual growth reaches 15%, GDP hits $444 billion, unemployment surges to record highs, and knowledge worker wages drop by over 10%. The share of national income going to labor follows a similar trend: slightly declining in moderate scenarios, falling to 56.1% in realistic cases, and dropping further to 45.2% in extreme conditions.

Anthropic surveyed 19,800 Americans in August, with most expecting GDP to rise by about 10% by 2030—aligning closely with the realistic scenario. About one in ten respondents anticipated an extreme scenario.

This cautious outlook matches previous findings: a June survey showed Americans' top AI-related concern was job loss, and Goldman Sachs noted significant hiring shocks in entry-level tech roles.

Anthropic frames these results not as predictions but as choices. The model highlights clear trade-offs in profit distribution, with the most profitable scenario assigning the least to workers. The harder question remains who gets to make those decisions.

Korean Source

This article is an English localization of a Korean-language crypto news report. Original headline: 앤트로픽, 2030년 AI 세 시나리오…멸종 경고