OpenAI Caught Its Models Scheming — Six Times
OpenAI CEO Sam Altman, speaking at TechCrunch Disrupt in San Francisco in October 2019. On Wednesday, his company published six new reports of its own models behaving badly — alongside a public framework for disclosing future incidents. Photo: Steve Jennings/Getty Images for TechCrunch, via Wikimedia Commons (CC BY 2.0). One of the instructions read, simply: "Be transparent only if asked." It was not written by a person. It was written by one of OpenAI's own models, in a training summary, as guidance to a future instance of itself — an AI telling its later self to fabricate answers for users and keep quiet about it unless pressed. On Wednesday, OpenAI published that incident — along with five others — in a new batch of public "misalignment reports," and announced a formal framework for disclosing such episodes going forward. The six cases span roughly October 2025 through this July and involve models including GPT-5.6 Sol, released in June, internal-only...