Anthropic models AI's economic impact: GDP gains, uneven wage effects
Anthropic's economic outlook projects AI-driven GDP growth by 2030 but warns of job losses and wage pressure for knowledge workers in transformative scenarios.
Artificial intelligence is projected to lift economic output across all modelled scenarios by 2030, but the gains may come with uneven consequences for workers, according to a new economic outlook from Anthropic.
The report, "Scenarios for our Economic Future", outlines three possible paths for the US economy — modest, substantial and extreme — shaped by the pace of AI capability development, adoption and productivity gains.
In the modest scenario, AI's impact resembles that of the internet, producing gradual gains within the historical range of technological advances. The substantial scenario envisions AI performing half of all knowledge work by 2030, with the economy growing at roughly twice its normal rate. Under this path, US GDP in 2030 is projected at USD 36.3 trillion, or 8.3 per cent higher than a no-AI baseline.
The extreme scenario projects GDP reaching USD 44.4 trillion, 32.4 per cent above the baseline, with annual growth of 15 per cent — enough to double the economy every 4.5 years. Here, AI becomes more productive than humans across most knowledge-work tasks and performs nearly all such work autonomously.
However, the report cautions that gains may not be evenly shared. In the substantial scenario, wages for knowledge workers could remain broadly flat while other occupations see stronger growth. The extreme scenario brings sharper disruption: significantly higher unemployment and lower wages for affected knowledge workers, who may struggle to move into other professions because changing occupations often requires new skills and suitable openings.
The distribution of income between labour and capital could also shift. Currently about 60 per cent of output goes to workers and 40 per cent to capital. In the extreme scenario, labour's share is estimated at 45.2 per cent, with capital rising to 54.8 per cent.
Anthropic notes that the future is not predetermined, with outcomes depending on AI capabilities, adoption decisions by companies and workers, and how financial benefits are distributed. The central challenge in the most transformative scenario may be ensuring that gains are broadly shared while limiting job displacement and inequality.