Evaluating the Impact of AI on the Labor Market: Current State of Affairs

A recent analysis of the U.S. labor market reveals no significant economy-wide disruption or accelerated changes in occupational mix attributable to generative AI since ChatGPT's introduction 33 months ago, undercutting widespread fears of immediate job displacement. The study indicates that shifts in the occupational landscape are comparable to previous technological advancements and often predate AI's public release, with measures of AI exposure and usage showing stable employment patterns. This suggests that while AI's long-term impact is still evolving and requires more comprehensive data, its immediate effects on the broader labor market have been gradual, consistent with historical technological adoption curves.

Analysis

Analysis of the US labor market 33 months post-ChatGPT's launch indicates no significant, economy-wide disruption attributable to generative AI. Contrary to widespread anxiety, the pace of change in the occupational mix is only marginally faster—by approximately one percentage point—than during previous technological shifts like the internet's adoption, and these trends largely predate AI's public release. While high-exposure sectors such as Information, Financial Activities, and Professional and Business Services are experiencing more rapid shifts in their job mix, the data suggests these are continuations of pre-existing trends rather than a direct consequence of AI. The study underscores the limitations of current data, highlighting a low correlation between OpenAI's theoretical 'exposure' metrics and Anthropic's 'usage' data, with the latter being heavily skewed towards specific professions like coding. This suggests that current fears of immediate, large-scale cognitive labor erosion are speculative, and the market impact is more likely to follow the historical precedent of a decades-long integration cycle rather than an abrupt shock.

Key Decisions for Investors

• Investors should moderate expectations for immediate, large-scale labor market disruption from AI, as the data points to a gradual, multi-decade transition similar to past technological waves rather than an abrupt economic shock.
• The low correlation between theoretical AI 'exposure' and actual 'usage' suggests investment theses based solely on identifying exposed industries are unreliable; a more effective strategy is to focus on companies demonstrating tangible productivity gains from AI integration.
• While economy-wide effects are muted, portfolio managers should monitor high-exposure sectors like Information and Professional Services for early signs of accelerating occupational shifts, as these may serve as leading indicators for broader market changes.