External signal·The New York Times·May 3, 2026·Ezra Klein·11 min read
Why the A.I. Job Apocalypse (Probably) Won't Happen
“Economists, I've found, are quite skeptical that mass joblessness is on the horizon.”
Summary
Klein opens with a March Quinnipiac poll showing 70% of Americans expect AI to reduce job opportunities, up from 56% a year earlier, and contrasts that anxiety with how skeptical economists are that mass joblessness is imminent. He sets AI leaders' warnings — including Dario Amodei's claim that up to half of entry-level white-collar jobs could dissolve within five years, and Mustafa Suleyman's faster timeline — against macro data showing stable unemployment. Leaning on economist Alex Imas's "What Will Be Scarce," he applies a Jevons-paradox argument: as AI makes some labor cheap, demand shifts toward whatever stays scarce. His closing warning is that a world displacing eight million workers may be harder to manage than one displacing eighty million.
Predictions for the future of work
Predicts no near-term mass-unemployment event, expecting labor demand to migrate toward non-automatable work rather than collapse. But warns the politically harder scenario is partial displacement — large enough to harm specific cohorts such as recent graduates and entry-level white-collar workers without being universal enough to force a policy response.
Originally published by The New York Times · May 3, 2026
Read the original at The New York Times