Geoffrey HintonDeep learning pioneer, Turing Award and Nobel laureate; known for the 2016 radiologists call and later AI-risk warnings.
People should stop training radiologists now. It's just completely obvious that within five years deep learning is going to do better than radiologists, because it's going to be able to get a lot more experience.
Machine Learning and the Market for Intelligence, Creative Destruction Lab, Toronto · Nov 24, 2016 · source · archived snapshot · hindsight clip
Admitted by the site's editor alone — Seeded by the site owner at launch; single-operator admission (ADR-0012).
Normalized claim
Deep learning will outperform human radiologists within five years, by late 2021.
Confidence
high (90%)
Deadline
Nov 30, 2021 · inferred
Resolution rule
Hit if by end of November 2021 AI systems were reading medical images better than radiologists across the job, such that radiologists were being replaced. Miss if radiologists remained essential and AI stayed an assistive tool; US radiologist employment and demand grew through 2021.
Crowd vote
No crowd votes were recorded.
Voting closed at the deadline.
More from Geoffrey Hinton
I now predict 5 to 20 years but without much confidence. We live in very uncertain times. It's possible that I am totally wrong about digital intelligence overtaking us. Nobody really knows which is why we should worry now.
OPEN