OpenAI launches embedded assessments with Apollo, METR, Transluce and Goodfire
MariusHobbhahn · x · 2026-11-17
Key facts:
- OpenAI announced embedded assessments: giving outside experts deep access to its internal models and systems to investigate alignment and test safeguards.
- Partners: Apollo Research, METR, Transluce AI, and Goodfire AI.
- The first phase, led by Apollo's Marius Hobbhahn's team (@lamaahmad), focuses on core alignment questions; more work on both substance and governance is expected in the coming months.
- The program adds independent scrutiny to how OpenAI identifies and manages risks in its systems.
Related event: OpenAI Launches Embedded Assessments with Four External Safety Labs(19 posts)→
More from Safety
- Flights, banks and Microsoft hit by mass outages; officials deny cyberattack — AaronBergman18 · 2026-11-17
- Kelsey Piper: AI labs have overwhelming commercial incentive to redact audit reports — NathanpmYoung · 2026-11-17
- Critical systems keep having unexplained outages that officials claim are unrelated — josh_bickett · 2026-11-17
- Pedro Domingos clashes with Neel Nanda: machines don't have values, says consensus — pmddomingos · 2026-11-17
- Anthropic's Recursive Self-Improvement Alarm Reportedly Triggered, Security Push Mandated — ForgotMyParachute · 2026-11-17
- CMU researcher warns aggregated chatbot analysis can leak sensitive research insights — niloofar_mire · 2026-11-17