Home / Series / Computerphile / Aired Order / Season 2025 / Episode 36

Sleeper Agents in Large Language Models

It's an older paper, but it checks out. Rob Miles discusses the problem of 'Sleeper Agents' - where LLMs could have hidden traits we don't know about until it's too late.

English
  • Originally Aired September 12, 2025
  • Runtime 13 minutes
  • Production Code wL22URoMZjo
  • Network YouTube
  • Created September 12, 2025 by
    Henke_tvdb
  • Modified September 12, 2025 by
    Henke_tvdb