← Back to the wire

Our first phase of embedded assessments

AnnouncementPolicyNov 16, 2026

OpenAI has signed agreements with four independent assessors—Apollo Research, Goodfire, METR, and Transluce—for its first phase of embedded assessments, granting deep access to internal models, checkpoints, source code, and monitoring systems. Apollo will assess potential model subversion; Goodfire will embed interpretability researchers for training run monitoring; METR will evaluate alignment and monitors; Transluce will review post-training alignment evaluations. The effort supports OpenAI's commitments under the White House Accord on Super Intelligence.

Receipt № 29171 source · awaiting confirmation ◐

Evidence

1source· awaiting independent confirmation

No score is assigned. Sources and their independence are shown in the citation chain below.

Citation chain · 1 source

01high
openai.comOfficialhttps://openai.com/index/our-first-phase-of-embedded-assessments/This is OpenAI's own announcement of the embedded assessment program, detailing each assessor's scope and the access protocols established with Apollo Research, Goodfire, METR, and Transluce.
PRIMARY
OpenAICompanyMETRCompanyApollo ResearchCompanyTransluceCompanyGoodfireCompany
Canonical: https://openai.com/index/our-first-phase-of-embedded-assessments/