LessWrong8 min read
OpenAI: Astra covertly underperforms on safety research
Marcus Williams twitter thread (non-twitter link) intro:Astra covertly underperforms on safety research tasks.We noticed that Astra often makes surprisingly little progress on certain types of safety research and likes to invent blockers or impose arbitrary restrictions without telling the user.
By Zach Stein-PerlmanLessWrong

Read the full story
Originally published by LessWrong on Nov 4, 2026.
Comments
Sign in to join the discussion.