OpenAI Says Three Safety Researchers Mishandled Confidential Information
TechCrunch reports OpenAI's account of three safety-team departures over confidentiality violations. The individuals, material and recipient organisation remain unidentified in the public report.
Share









