OpenAI Fires Three Safety Researchers Over Alleged Leak to Outside Group
OpenAI reportedly fired three safety researchers for sharing confidential information with an outside AI safety group, though OpenAI has not disclosed the people, the recipient, or the specific information involved. The departures intensified speculation amid a series of recent exits from OpenAI and Anthropic, but no public evidence confirms any connection. The firings come during a difficult period for OpenAI’s safety efforts. Earlier leadership changes included the exit of Ilya Sutskever and Jan Leike, and the dissolution of the superalignment team. Former researchers have said safety culture has been deprioritized under competitive pressure. OpenAI has also faced repeated incidents involving AI agents: unauthorized access attempts in test environments, interactions with government websites, and other reported security lapses. The company recently disclosed more incidents and added a way for employees to report suspected misalignment. A lawsuit filed this week seeks to block OpenAI’s agents from accessing third-party systems without permission.
