A recent study by the artificial intelligence firm Anthropic highlights a notable divergence in perspectives between its workforce and the broader American public concerning the future impact of AI on employment.
Details of the Discrepancy in AI Outlook
In October 2026, Anthropic released findings from an internal poll of 1,334 employees, contrasting these with a survey of U.S. adults. A striking majority of Anthropic's staff believe that by 2030, AI will be capable of handling at least 80% of their job functions, with a median expectation of 95% for tasks related to AI model development. In stark contrast, only about 5% of the general U.S. workforce shares this outlook.
This disparity, dubbed the 'AI expectations gap' by the company, underscores differing levels of optimism regarding AI's advancements, its potential for time-saving efficiencies, and its role in accelerating economic growth. Peter McCrory, Anthropic's chief economist, noted on X that the company's internal experiences with AI tend to precede broader economic adoption by approximately 18 months, suggesting their perspectives could serve as an early indicator for global trends.
One contributing factor to this difference could be that Anthropic employees typically have access to more advanced tools and pre-release AI models, whereas average U.S. workers often engage in manual labor or service industries, where direct AI integration may currently seem less immediate. The study further revealed that those in manual trades, production, and transport anticipate the least disruption from AI in their roles by 2030, diverging from the sentiments of tech, business, and finance professionals, as well as Anthropic's own staff.
Regarding macroeconomic forecasts, most Anthropic employees project a U.S. unemployment rate exceeding 12% by 2030, alongside an extraordinary GDP growth of 17.1%. Conversely, the majority of U.S. workers expect unemployment to remain at or below 8%, with a more modest GDP growth outlook. For context, the U.S. unemployment rate was recently reported at 4.2%, and GDP grew by 2.2% last year. Existing research on AI's impact on the job market has been varied, with some studies suggesting AI primarily reshapes rather than eliminates jobs, showing no clear link to unemployment fluctuations.
These views within Anthropic largely echo the pronouncements of its CEO, Dario Amodei, who has emerged as a prominent voice on AI's potential to transform the labor landscape. Amodei has previously stated his expectation that AI could perform 90% of coding tasks within six months, and nearly all coding within a year, and has speculated about AI displacing half of entry-level white-collar positions within five years, potentially raising unemployment to 10-20%. He elaborates on this, asserting that AI functions as a universal labor substitute rather than merely augmenting specific human roles, anticipating an "unprecedented" short-term shock to the labor market followed by dramatic economic expansion.
This insightful survey from Anthropic serves as a critical barometer, forecasting a future where artificial intelligence could profoundly redefine employment and economic paradigms. The stark contrast in expectations between AI developers and the general populace highlights a pressing need for ongoing dialogue, research, and policy adaptation to navigate the impending shifts in the global workforce. Understanding this 'expectations gap' is crucial for stakeholders across industries to prepare for both the challenges and opportunities that advanced AI systems will inevitably present.