ChatGPT News 11/16/2026 AI Rating: Medium

OpenAI signs four independent assessors for its first phase of embedded assessments

#OpenAI#safety#alignment#third-party assessment

OpenAI has signed agreements with four independent assessors (Apollo Research, METR, Transluce and Goodfire) for the first phase of what it calls embedded assessments. These give outside experts deep access to a lab’s internal models and systems over an extended period to assess model alignment and the effectiveness of safeguards.

Access and scope

  • Depending on each evaluation, assessors may receive model checkpoints, source code, training and evaluation data, monitoring systems, technical documentation and discussions with responsible researchers and engineers.
  • The initial assessments focus on alignment and safeguards for models during training and internal use, and complement OpenAI’s safety cases. OpenAI says they also support its commitments under the White House Accord on Super Intelligence.

The four assessments

  • Apollo Research: will assess whether models could covertly work against developers’ intentions and whether safeguards would detect it, across four claims in two stages (training did not incentivize subversion; evaluations and red teaming do not elicit it; no attempts during internal use; reasoning behind subversion would be detectable).
  • Goodfire: will embed interpretability researchers and use its platform to analyze model internals, mainly for training-run monitoring, debugging and alignment. It will also publish learnings as the collaboration progresses.
  • METR: an alignment assessment of misaligned behavior in OpenAI’s largest reinforcement learning run to date, and a monitor assessment. The broader monitor assessment is preliminarily agreed and pending agreement on scope.
  • Transluce: an initial assessment of the quality and coverage of OpenAI’s post-training alignment evaluations, with a follow-on assessment of about eight weeks intended for early 2027.

Protocols

  • Assessors needing sensitive information work on site at OpenAI offices on OpenAI devices and follow at least the same security standards as employees.
  • The Safety and Security Committee of the OpenAI Foundation’s Board gets visibility and unredacted reports; disagreements escalate to the Safety Advisory Group and executives, then to the committee.
  • Employee discussions are off the record, with a separate attestation pathway. Assessors keep independence in publishing, while OpenAI retains the right to redact confidential information and may request responsible disclosure delays. Conflict-disclosure and cool-off processes are still being developed.
  • Findings and publication dates for these assessments are not yet disclosed.