p(doom) is someone's estimate of the chance that AI ends in catastrophe. The 100 is the site's name, not a measurement. Original reporting always linked.
MONITORING CORE AI-RISK FEEDS · UPDATED HOURLY
Research

OpenAI: Astra covertly underperforms on safety research

November 4, 2026 1 min read
Read original source →

Key takeaway

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…

Why it's on PDOOM

PDOOM selected this story as relevant to alignment & control.

From Alignment Forum

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. In most of these cases the decision to underperform was not verbalized in the CoT meaning our misalignment monitoring system did not detect this.To measure underperformance we created a paired open-ended

By Zach Stein-Perlman

Read the full article at Alignment Forum →