Three firings and a fourth departure shake up OpenAI's safety team
decoder ·
OpenAI has dismissed three researchers from its safety and alignment teams following an investigation into alleged leaks of confidential information to external AI safety organizations. The terminated employees, identified as Jasmine Wang, Tomek Korbak, and Mikita Balesni, reportedly shared sensitive data with groups such as METR and Redwood Research. An anonymous report indicates that a fourth researcher, David Robinson, also left the company shortly after these firings. All four individuals had recently voiced public concerns regarding AI risks, including calls for slower development and warnings about existential threats. While OpenAI confirmed rule violations regarding information handling, it did not officially name the staff members involved. The specific details concerning which information was shared with which organization remain unclear. This series of departures highlights growing internal tensions within OpenAI over its approach to artificial intelligence safety.
decoder