Anthropic economist contradicts CEO's AI job-loss warnings

Peter McCrory, Anthropic's head of economics, published an X essay this week synthesizing 18 months of the company's internal economic research to argue that AI has caused no material rise in US unemployment. The essay lands directly in tension with CEO Dario Amodei's repeated public warnings of an imminent labor crisis. In May 2025, Amodei told Axios that AI could wipe out half of all entry-level white-collar jobs and spike unemployment to 10%-20% within one to five years, urging companies and policymakers to stop "sugarcoating" the risk. He doubled down in a January 2026 essay titled "The Adolescence of Technology," warning that AI functions as a "general labor substitute for humans" that will displace work from lower skill tiers to upper tiers, potentially creating a lasting underclass of unemployed or very-low-wage workers.

Amodei's framing has shifted significantly over the past year. By May 2026, he moderated somewhat, reframing automation as a multiplier of output and invoking the Jevons paradox: "If you automate 90% of the job, then everyone does the 10% of the job," he said, explaining that "the 10% kind of expands to be 100% of what people do and kind of 10-times their productivity." But the following month he ratcheted up again, arguing in June that significant, enduring job loss might be "an intrinsic property of the technology" itself and calling for government responses including wage insurance and universal basic income.

McCrory's data tells a different story. The unemployment rate stood at 4.2% in June—a level the Federal Reserve associates with full employment—while job openings roughly matched the number of unemployed workers and prime-age employment sat near multi-decade highs. Most pointedly, McCrory's updated analysis using recent Bureau of Labor Statistics data shows no relative deterioration in unemployment among workers whose jobs contain a large share of tasks that Claude automates, compared with workers in less-exposed roles. "I don't expect unemployment to be noticeably higher a year from now—at least not because of AI," he wrote.

McCrory attributes the gap to what he calls AI's "stubbornly jagged" capability profile: no job in the Labor Department's O*NET taxonomy has all of its tasks handled by Claude, and complex work still depends on human oversight to direct systems and catch their errors. He pointed to evidence that Claude usage correlates with users acting as "thought partners" rather than replacements, and that people with more domain expertise succeed more often and recover better when the AI stumbles. McCrory's data most directly undercuts Amodei's "general labor substitute" theory: if entry-level consultants, lawyers, and financial analysts were being systematically substituted out at the pace Amodei described to 60 Minutes in November 2025, occupation-level unemployment data should already show divergence, yet it does not.

Yet the article notes an unresolved tension remains. McCrory concedes that hiring for young workers in highly AI-exposed roles has softened over the past year, consistent with Stanford research on "canaries in the coal mine," and flags Bureau of Labor Statistics projections of slower growth through 2034 for occupations like technical writers, data entry workers, and customer support reps. Both Amodei and McCrory agree that early-career, entry-level workers in highly exposed roles are the most vulnerable group right now. McCrory also conceded that the future is uncertain: if AI begins automating innovation itself through recursive self-improvement, standard economic models allow for a scenario resembling the "singularity" that Amodei fears—just not, in McCrory's read, on the near-term timeline or scale his boss has publicly forecast. The company thus remains publicly straddled between two positions: an in-house data scientist insisting the disruption hasn't shown up yet, and a chief executive who has said repeatedly that disruption is coming fast and could warrant policy responses on the scale of universal basic income.

The divergence between McCrory and Amodei reflects a fundamental tension in how to interpret early signals of AI's labor impact. Amodei has shifted his framing repeatedly over the past 18 months—from a catastrophic near-term scenario of mass white-collar job elimination to a more recent invocation of the Jevons paradox, which posits that automation expands productivity and output rather than simply erasing jobs. McCrory's data appears to support this later, more moderate framing: unemployment is near historic lows, job openings match joblessness, and workers in roles heavy with automatable tasks show no relative deterioration in employment compared to less-exposed roles.

However, McCrory acknowledges cracks in the foundation. Hiring for young workers in AI-exposed roles has softened over the past year, and the Bureau of Labor Statistics projects slower growth through 2034 for occupations like technical writers, data entry workers, and customer support reps. This pattern—the most vulnerable cohort feeling pressure first—is consistent with both the catastrophe scenario and the gradual adjustment scenario. McCrory's own framework suggests that AI amplifies the productivity of skilled workers but that transition time is required, yet Amodei has repeatedly argued AI is moving faster than any past general-purpose technology, which is precisely the condition under which smooth market rebalancing could break down. McCrory's implicit concession—that recursive self-improvement could eventually produce the "singularity" Amodei fears—leaves open the possibility that Amodei's timeline is optimistic but his direction is correct.

No comments yet. Be the first to share your thoughts!