Yale economist says AGI won’t automate most jobs — because they’re not worth the trouble

The claim

It has been reported that a Yale economist argued AGI won’t lead to wholesale job automation because, in many cases, automating a job simply isn’t worth the trouble. The claim, surfaced on a Reddit technology thread, paints a picture of an economy littered with low-value, fragmented, context-heavy tasks that are expensive to automate and awkward to integrate into firm operations. In short: just because an AI can do something in the lab doesn’t mean it makes economic sense to replace a human on the factory floor, in the storefront, or at the front desk.

Why it matters

If true, this reframes the automation panic. Fewer white-collar apocalypses, perhaps. But don’t breathe easy yet. The argument pushes policymakers and businesses to think in dollars and frictions — not fantasy. Which jobs get automated will depend on transaction costs, deployment headaches, and the value captured by firms, not just technical capability. That’s a useful corrective to breathless headlines that treat every new demo as a harbinger of mass unemployment.

Caveats and reactions

Allegedly, the economist’s stance has provoked pushback from researchers who note that automation historically arrives in waves and often in unexpected places. Reddit isn’t peer review; the original post is a summary, not a journal article. Experts point out that high-value, routine tasks already show heavy AI uptake, and infrastructure improvements can change the calculus quickly. So the emotional moment here — relief that AGI won’t wreck livelihoods overnight — is understandable, but it should be tempered by vigilance. After all, slow burns can still burn down the house.

Sources: reddit