A 2026 study by Johnston and Makridis found higher AI exposure across US industries was linked to 7% higher output between 2020 and 2024, using employment, GDP and wage data.
The same one-standard-deviation increase in AI exposure was associated with 3.9% higher employment and a 4.8% rise in employers' wage bills, suggesting productivity gains coincided with job and pay growth.
The analysis tracks industrial AI adoption from 2021 and generative AI's spread after ChatGPT's 2023 launch, framing the results as early macro evidence for Jevons' Paradox rather than mass white-collar job loss.
Researchers cautioned the data is still early, but the findings add to recent survey evidence that AI proliferation has so far not produced broad job destruction in the US.