Bryan Cantrill: AI doomers like Hinton are wrong because they overreach their expertise

rms80 · x · 2026-09-29

Bryan Cantrill's essay "Fool's Expertise" critiques the Ezra Klein interview with Nvidia CEO Jensen Huang, calling it a Rorschach test for AI doomerism, and argues Huang missed the real counterpoint: Hinton-style predictions fail not because the future is unclear but because they exceed the speaker's domain expertise. The 2016 claim that radiologists would vanish by 2021 ignored what radiologists actually do — image interpretation is only part of the job, and better tools amplify their value.

Original post →

More from AGI Musings

AGI Musings channel →