Anthropic Maps Three AI Futures: From Internet-Scale to 18% Unemployment

Three scenarios for 2030

Anthropic has released an economic model projecting how AI could reshape the U.S. economy by 2030, outlining three distinct futures that range from modest productivity gains to transformative disruption. Published September 9, the analysis moves beyond binary "will AI take my job" questions to examine which specific tasks automation might handle across the workforce.

The company's framework examines two critical variables: what percentage of work tasks AI can perform and whether displaced workers find new employment. Using a nurse's shift as an example, Anthropic notes AI cannot bathe patients but could draft discharge instructions or manage supply orders autonomously. The model then scales this task-level analysis across the entire economy.

The mild scenario

In Anthropic's first projection, AI touches roughly 4% of economic tasks by 2030, primarily assisting rather than replacing human workers. Annual GDP growth would accelerate modestly from 2% to 2.4%, while unemployment would tick up marginally from 3.8% to 3.9%. Anthropic characterizes this outcome as comparable to the internet's economic impact.

The middle path

The second scenario envisions AI handling 12% of work tasks, with most applications replacing rather than augmenting workers. Economic growth would reach 5.4% annually, and unemployment would climb to 4.6%. This represents meaningful disruption but remains within historical precedent for technological transitions.

The extreme case

The third scenario departs dramatically from incremental change. AI would impact 30% of all economic tasks—representing about half of knowledge work—and perform nearly all of it independently without generating replacement jobs. This outcome would require AI systems capable of self-improvement, according to Anthropic.

Under this projection, GDP growth would hit 15.4% annually, fast enough to double the economy every 4.5 years. However, overall unemployment could reach 11.9%, with knowledge workers facing a 17.9% jobless rate. The authors note these "consequences go well beyond any event in history."

Why it matters

Across all three scenarios, including the mildest, Anthropic's model shows wealth shifting toward asset owners rather than workers—a trend that would intensify existing inequality concerns. For business leaders, the analysis suggests planning for a wide range of outcomes rather than a single predetermined future. The model also highlights that AI's economic impact will be determined not just by technical capability but by whether the economy generates new roles for displaced workers—a factor shaped by policy, investment, and institutional adaptation rather than technology alone.

These projections were first reported by Yahoo Finance based on Anthropic's published economic model.