U.S. employment data so far show a modest change in some AI-exposed occupations, while Bridgewater Associates has warned that AI could displace as much as 18% of U.S. jobs within five years. That figure is a forecast, not a measure of job losses already caused by AI.
Jensen laid out the estimate in comments reported by Bloomberg, which also notes that Jensen was one of the earliest backers of both OpenAI and Anthropic — giving him a front-row seat to the technology's rapid rise well before it became a Wall Street obsession. Bloomberg reports he recently published an op-ed in the New York Times proposing a “token tax” as one way to help cushion the job losses AI might cause, a policy idea aimed at the same disruption he says is barreling toward the labor market.
The warning carries extra weight because it comes from inside a firm that is racing to build the very technology Jensen warns about. Bridgewater's AIA Labs describes itself as the company's dedicated artificial intelligence research and investment lab, spearheaded by Jensen, according to Bridgewater. The lab says it is building what it calls an artificial investor, or AIA, aiming to meet or exceed expert human performance across the full range of tasks that investors perform.
AI Tools Already Managing Billions
Bridgewater's own account of its AI systems suggests the technology has moved well past the experimental stage. The firm says its AIA systems have been tested with real capital and now manage billions of dollars while generating alpha, per the same Bridgewater page. The company also claims its AIA Forecaster is the first publicly documented AI system to verifiably match the performance of expert human forecasters at scale.
Bridgewater frames the effort as a partnership rather than a replacement, at least internally. The firm's AIA Labs exists in deep partnership with its flagship Pure Alpha strategy, the company says, and it maintains that the boundary between human and artificial intelligence in investment research will eventually dissolve. Bridgewater adds that AIA Labs is committed to studying how artificial intelligence is reshaping markets, the broader economy and society.
What the Broader Labor Data Shows
Outside Bridgewater's walls, the real-world evidence on AI and jobs is more mixed than Jensen's stark figure might suggest. Employment in 18 occupations the Bureau of Labor Statistics identifies as exposed to artificial intelligence fell 0.2% between May 2024 and May 2025, according to a separate Bloomberg analysis, even as overall U.S. employment rose 0.8% over the same period. Those AI-exposed roles account for roughly 10 million jobs, the outlet reports, with losses led by customer service representatives and certain types of secretaries and salespeople.
A Closer Look at Customer-Service Employment
A U.S. Bureau of Labor Statistics Current Population Survey series displayed by the Federal Reserve Bank of St. Louis’ FRED counted 1,928 thousand full-time wage-and-salary workers in customer service representative occupations in 2025. That is an employment level, not evidence that AI caused job losses, according to the FRED series.
Federal researchers studying the broader picture have found a more muted effect so far. Census Bureau researchers, using Business Trends and Outlook Survey data, found that AI use is prevalent in large firms and knowledge-intensive sectors, that it generally augments rather than replaces tasks, and that labor declines tied to AI adoption remain rare, according to the Census Bureau. Separately, a Federal Reserve note cites equal-weighted adoption rates of 69.4% for AI generally and 46.4% for large language models specifically among its surveyed sample, according to the Board of Governors of the Federal Reserve System.
How Regulators Are Tracking the Shift
The federal government's own forecasting tools are still catching up to the pace of the debate. The Bureau of Labor Statistics says its employment projections are prepared annually for a 10-year horizon and cover roughly 800 detailed occupations across about 300 detailed industries, according to the agency. The BLS has also built an occupational-utilization table that it says provides information about factors, including AI, that may impact future demand for a given occupation.
Reports differ on who delivered the 18% warning: Bloomberg identifies Jensen, while an October 1 report carried by TradingView attributes the quoted warning to Bridgewater CEO Nir Bar Dea. The difference underscores the need to treat the figure as a reported forecast, not a settled measure of AI's labor-market effects.
