[Linkpost] “AI misbehaviour in the wild from Andon Labs’ Safety Report” by Lukas Petersson | LessWrong (30+ Karma) | Podwise