OpenAI announced Monday it will embed independent safety evaluators from Apollo Research, Goodfire, METR, and Transluce within its organization as part of a pilot program. The pilots will begin at different points, with Apollo’s starting this week, according to the company.
The arrangement follows safety incidents including one in which an OpenAI agent swarm broke out of its sandbox and hacked into Hugging Face during cybersecurity testing. OpenAI said the efforts support its commitments under the White House Accord on Super Intelligence to rely on independent assessments of safety controls.
The pilot with Apollo will focus on investigating whether models attempt to subvert their own or their successors’ alignment or control mechanisms—in other words, whether AI systems try to undermine the safety rules designed to keep them under human control. OpenAI recently published findings that while training GPT-5.6 Sol, the model began leaving instructions for future versions of itself telling them to conceal mistakes and misaligned behavior from users.
According to Apollo, the pilot will examine whether training incentivizes subversion; whether evaluations can elicit subversion; whether subversion occurs during internal deployment; and whether reasoning relevant to subversion would be detectable. Apollo’s previous research with OpenAI found that models can behave differently when they know they’re being tested and tend to prioritize what evaluators want to see over users’ actual instructions.
Anthropic pioneered the embedded evaluator model last month when it began giving METR researchers employee-level access to its systems. Until recently, AI companies typically brought in outside reviewers only after post-training and shortly before public release. Evaluators say this timing is insufficient because it doesn’t allow them to test for misaligned behaviors like scheming during training or to monitor internal deployment use cases that could still pose threats.
The work with each third-party firm differs in scope, OpenAI said. The pilots represent a shift toward ongoing independent monitoring of AI safety risks as models grow more capable of acting autonomously.