OpenAI revises framework for severe frontier-model risks
OpenAI released Preparedness Framework version 2 with tracked catastrophic-risk categories, safeguards reports, Safety Advisory Group review and new research categories including long-range autonomy, sandbagging and safeguard undermining.
AWAY FROM DOOM34confidence 72/100
Why it moved the index
The revised framework makes severe-risk evaluation and safeguards review more operational, while its provider-controlled and revisable nature limits confidence in realized control strength.
0 comments · 0 votes
Sign in to join the discussion →
No comments yet. Start the discussion.
AUDIT TRAIL
Assessment history
- R1Away 34 · confidence 72
Material successor to the durable 2023 framework event with distinct categories, governance and deployment criteria.
12 Aug 2026