Anthropic Economist Finds No Evidence of 'AI Jobs Apocalypse' in US Data

The predicted mass displacement of white-collar workers by artificial intelligence has not yet materialized in the United States labor market. An analysis of 18 months of data reveals that the feared 'AI jobs apocalypse' remains a theoretical risk rather than a current reality.

Peter McCrory, Head of Economics at Anthropic, reviewed Bureau of Labor Statistics and internal data to track employment trends since late 2022. His findings show no systematic increase in unemployment for workers in occupations highly exposed to AI. Specifically, professionals in high-exposure roles—including lawyers, consultants, and analysts—are performing no worse in the job market than those in roles with low AI exposure. As noted in the Anthropic report, "We find no systematic increase in unemployment for highly exposed workers since late 2022."

The Deployment Gap

This data creates a striking contrast within Anthropic itself. The company's CEO, Dario Amodei, has previously warned that AI possesses the potential to wipe out a significant portion of entry-level white-collar positions. However, the discrepancy between Amodei's warnings and McCrory's data is explained by the gap between technical feasibility and actual implementation. While AI is technically capable of performing many tasks, its actual deployment within the workplace remains a small fraction of its total potential.

The Shift to Judgment

These findings suggest that AI is currently functioning as a collaborative tool rather than a direct labor substitute. The economic implication is a shift in the value of human labor: the "production layer" of work—the routine generation of content or data—is becoming cheaper and more automated. In contrast, the "judgment layer" is increasing in value. This layer encompasses human expertise, strategic oversight, and the ability to provide direction to AI systems, making human discretion more critical as the cost of production drops.

What to Watch

While the aggregate data shows stability, the long-term trajectory remains uncertain. The primary question for economists is whether the current stability is a permanent shift toward collaboration or merely a delay caused by slow corporate adoption. As companies move from experimenting with AI to fully integrating it into their operational workflows, the pressure on entry-level roles may intensify. For now, the data indicates that the feared collapse of the white-collar job market has not begun.