OpenAI terminated three safety researchers for violating the company's policies on accessing and handling sensitive information, according to The Wall Street Journal. An OpenAI spokesperson confirmed the departures, stating: "We have parted ways with three individuals for violating our policies on accessing and handling sensitive company information. Our investigation confirmed that these individuals accessed and handled sensitive company information in violation of our policies."
The three researchers worked on OpenAI's safety team. The company did not disclose the specific nature of the sensitive information that was mishandled or the details of how the violations occurred. OpenAI's investigation determined that the departures were warranted based on policy breaches related to information access and handling protocols.
This incident highlights internal security practices at major AI companies regarding employee access to proprietary and sensitive data. OpenAI maintains strict policies governing how staff members interact with company information, particularly within safety-critical divisions. The company enforced these policies through investigation and termination of employment.
The Wall Street Journal first reported the story, though OpenAI later confirmed the basic facts through its official statement. The company did not elaborate on whether the leaked information was disclosed externally or remained within internal channels, nor did it specify consequences beyond employment termination.