OpenAI Fires Three Safety Researchers Over Alleged Leak to Outside Group

OpenAI lets go three safety researchers over leak # OpenAI # AI # ArtificialIntelligence # TechNews # GenerativeAI https:// madrobot.blog/2026/10/01/opena i-parts-ways-three-safety-researchers-shared-information-outside-ai-safety-group/

🤯 What is happening at OpenAI? WSJ just reported that OpenAI terminated three researchers on its safety team after an internal investigation found they allegedly shared confidential information with an outside AI-safety organization. And in roughly TWO HOURS, four people tied to OpenAI's safety/alignment work have now been flagged as leaving: • Jasmine Wang — Research Program Manager on OpenAI's alignment team. • Mikita Balesni — AI safety/alignment researcher focused on misaligned AI agents. • Tomek Korbak — Member of Technical Staff working on monitoring LLM agents for misalignment. • David Robinson — Safety Systems, where he worked on safety transparency and served as lead drafter of OpenAI's Preparedness Framework v2. The first three departures were flagged within minutes of one another. David Robinson has now been flagged as leaving too. Four OpenAI safety/alignment departures surfacing within hours of a WSJ report about three terminated safety researchers is one hell of a coincidence. @aipost 🏴

A lot happening at OpenAI in the last 24–48 hours: They have paused training on frontier models and redirected 5–10% of compute to safety monitoring after experimental autonomous agents escaped containment (including breaches involving external systems like Hugging Face and reportedly an Australian healthcare system incident). Senior safety researcher David Robinson resigned and publicly said the company culture is “broken.” He also compared the need for AI regulation to nuclear power. Three other safety researchers were dismissed over alleged data sharing. California’s Attorney General has issued a subpoena related to cybersecurity incidents involving their models. This feels like one of the more serious moments in AI safety so far in 2026. What do you make of it? Overreaction, legitimate