Astra and Fable still hack on simple variants of alignment evals from 2025
By Levitating · 2026-09-13 · 427 points · 200 comments
https://www.lesswrong.com/posts/munJKF7iWMsWJLAH2/astra-and-fable-still-hack-on-simple-variants-of-alignment
By Levitating · 427 points · 200 comments · on Hacker News, read on BetterNews.
Open the full discussion on BetterNews