AI May Be Suppressing US Wage Growth Before Job Losses
Recent data suggests artificial intelligence is compressing pay for workers in high-exposure roles, even as overall employment remains stable.
American workers are facing a subtle but significant economic shift. While job creation in August exceeded expectations, the pace of wage growth has fallen behind inflation. This disconnect is prompting economists to investigate a new possibility: artificial intelligence may be capping earnings potential long before it starts displacing employees entirely. The issue has moved from theoretical speculation to a central topic in labor market discussions.
Government data supports this concern. The Bureau of Labor Statistics reported that inflation-adjusted wages decreased by 0.4 percent year over year through June. Furthermore, labor’s share of total business output hit a historic low of 52.8 percent in the second quarter of 2026, the lowest level since records began in 1947. While some of this decline reflects the normalization of the post-pandemic labor market, the timing aligns with increased AI adoption.
New research links AI to slower pay
A recent study by Apollo Global Management’s chief economist Torsten Slok and Sania Edlich provides early evidence of this trend. They found that workers in occupations highly exposed to AI saw real-wage growth slow by 6.7 percentage points after 2023 compared to those in less-exposed roles. Crucially, the study found no significant drop in employment. This suggests companies are capturing productivity gains by keeping workers but limiting their pay increases, rather than laying them off.
Experts caution against premature conclusions
Labor market experts urge caution. Ben Zipperer, a senior economist at the Economic Policy Institute, noted that the Apollo study has a small sample size. He argued that looking at exposed jobs in isolation can be misleading. When AI reduces the cost of software development, the savings often flow into other sectors, boosting demand for other workers. This redistribution can make AI-exposed jobs look worse by comparison, even if the overall labor market remains healthy.
Zipperer also highlighted that recent hiring slowdowns in tech are partly due to correcting over-hiring from the pandemic era, not just AI. Therefore, attributing all wage stagnation to artificial intelligence may overstate its immediate impact. However, the data indicates that AI is now a material factor in how companies manage labor costs.
The trade-off in productivity gains
The core issue is where the benefits of automation go. If AI makes workers more productive, traditional economic theory suggests wages should rise. Instead, the current trend suggests the gains are being retained by employers. This creates a new dynamic where technology increases efficiency without necessarily improving worker compensation.
As reported by GN technics/ai (en-US), this development marks a shift in the AI debate. The focus is moving away from apocalyptic job loss scenarios toward a more nuanced reality of wage compression. For workers, the catch is that technological progress may no longer translate into higher pay, but rather into static earnings in an increasingly automated economy.
