Roman Yampolskiy: Dangers of Superintelligent AI | Lex Fridman Podcast #431

Lex Fridman · 135:39

Roman Yampolskiy tells Lex Fridman that a general superintelligence is not a safer, smarter tool but an unsolvable control problem: unlike a stolen password, civilization gets one shot, and he sees no durable good out...

Read the full summary on tuber

Redirecting...