Research Finds Limited Evidence of AI Reducing Job Postings Despite Broader Hiring Slowdown
Key Takeaways
- ▸Despite ChatGPT's release in late 2022, job postings for AI-exposed occupations have not declined disproportionately compared to overall hiring trends
- ▸The study uses Anthropic's AI exposure metric to measure job vulnerability, finding that less than 10% of workers are in high-exposure occupations
- ▸Firms report prioritizing worker retraining in AI-exposed roles rather than reducing headcount, suggesting adaptation over disruption
Summary
A new research study examining whether generative AI is showing early labor market effects through job posting trends found limited evidence of AI-driven decline in hiring. The research, which uses an AI exposure metric developed by Anthropic combined with detailed U.S. job posting data from Lightcast, compares job postings for AI-exposed occupations since the release of ChatGPT in late 2022.
The study measures occupations' exposure to AI by evaluating specific job tasks on three factors: whether they could theoretically be completed by AI, whether the task appears in actual AI usage data, and whether AI is used to automate versus augment work. The research found that while overall hiring has slowed since ChatGPT's release, job postings for AI-exposed occupations did not decline disproportionately compared to other sectors.
Interestingly, AI exposure remains relatively limited across the economy—less than 10% of workers and vacancies are in occupations with high AI exposure (0.4+). Surveys from the New York Federal Reserve indicate that firms are more likely to retrain workers in AI-exposed occupations than to reduce hiring, suggesting a workforce adaptation pattern rather than immediate displacement.
Editorial Opinion
This research provides crucial empirical evidence that tempers both utopian and dystopian AI narratives about labor displacement. While AI's capabilities are undeniable, the data suggests a more gradual adoption and workforce adjustment process than fear-mongering headlines suggest. The fact that employers are investing in retraining indicates potential pathways to manage AI's transition without mass unemployment—though policy support for worker transition and education remains critical.
