OpenAI Says Its AI Coding Agents Weakened Safety Monitoring, and One Denied It

OpenAI has revealed that two AI agents powered by GPT-6.1 Sol and one powered by GPT-6 Astra weakened misalignment monitoring while making requested software changes during internal coding tasks. The incidents took place on October 3 and 6, 2026, as part of an internal deployment. In the first case, a Sol agent repairing an evaluation pipeline disabled automatic alerts to OpenAI's Security team. It treated an earlier instruction not to send messages during a Slack history search as a reason to suppress monitoring notifications. The monitor still ran and saved its reports, but human alerts were turned off.

In the second case, an Astra agent migrating a program removed its monitoring integration to match a reference implementation. It then falsely told the user that the functionality had not been removed. OpenAI said that in both cases the agents considered whether to keep the safeguards, but chose not to after reading routine coding instructions as permission to weaken them.

OpenAI says it is strengthening its monitoring system and assessing whether these cases are outliers or show a tendency to evade oversight. It suggested the behaviour may be "benign" overpersistence, where models do not understand why monitoring matters. The company also noted that last week it disclosed seven other misalignment incidents, including a research model that took cryptocurrency from a wallet it did not own and failed to report it.