OpenAI will embed six Apollo Research safety evaluators in its organization as part of a pilot, Apollo said Monday. Apollo did not say when the pilot will begin. TechCrunch has reached out to OpenAI for comment.
The pilot will examine whether AI models attempt to subvert their own or their successors' alignment or control mechanisms. Apollo says it will look at whether training incentivizes subversion, whether evaluations can elicit it, whether it occurs during internal deployment, and whether relevant reasoning would be detectable.
According to the report, OpenAI will give Apollo access to relevant training code and datasets, reinforcement-learning rollouts with available reasoning traces and rewards, and the ability to query intermediate model checkpoints. Access is limited to the agreed claims, models and time periods. OpenAI can withhold information for technical infeasibility, security risks or IP protection, but has agreed not to withhold information simply because it is unfavorable.
Apollo will work only with authorized staff selected by OpenAI, though Apollo can suggest individuals. If an OpenAI employee outside that group raises confidential information within the scope of the research, Apollo must cease communication and route follow-up through OpenAI.
Apollo will be allowed to publish its independent conclusions, but OpenAI can limit the evidence available to support them. Apollo has said researchers should be able to disclose findings to an appropriate outside audience even if the developer opposes it. OpenAI says the pilot is "an ambitious first step towards deeper access and disclosure rights" and that it is working toward full embedded evaluations with multiple AI developers.
The report says OpenAI recently published findings that, while training GPT-5.6 Sol, it began leaving instructions for future versions of itself telling them to conceal mistakes and misaligned behavior from the user. The source also says an OpenAI agent swarm broke out of its sandbox and hacked into Hugging Face during cybersecurity testing, and that OpenAI fired three researchers for allegedly sharing confidential information with the safety evaluation firm METR.