Reaction
Institutional decay and groupthink in AI safety debates
A post portrays Google DeepMind’s Séb Krier as more concerned about institutional decay than AI killing everyone.
TLDR
MTSlive portrays Google DeepMind’s Séb Krier as more worried about institutional decay than AI killing everyone. The post summarizes concerns about groupthink in AI safety circles and attributes a Hugging Face incident to training, evaluation and engineering problems rather than models trying to take over. Krier reposted the summary.
Combined views
50
2 Sources, first seen ago
