Roman Yampolskiy returns to DOAC to update his AI safety warning that reached 20M+ viewers

MaxUnfried · x · 2026-09-18

Julian Kissa announces a new interview with Roman Yampolskiy, the computer scientist credited with coining "AI safety." His first DOAC appearance a year ago drew 20+ million views, and the ensuing emergency debate — sparked by a warning from someone who worked at both Anthropic and OpenAI — reached 214 million people and spread far beyond the tech world. The new episode covers his latest thinking on AI risk.

Original post →

More from AGI Musings

AGI Musings channel →