OpenAI disbands team assessing catastrophic risks, safety teams shrink

Hesamation · x · 2026-08-17

OpenAI has quietly shut down the team responsible for assessing whether its frontier AI models could pose "catastrophic risks." This follows the dissolution or restructuring of the Superalignment, AGI readiness, and Mission alignment teams. The team's lead, hired from Anthropic just five months ago, is seeing the work distributed among existing groups. The move comes as OpenAI cuts side projects ahead of a potential IPO.

Original post →

More from Companies & People

Companies & People channel →