Miles Brundage: OpenAI's security issues are a culture and leadership problem, not a sandbox problem

Miles_Brundage · x · 2026-09-29

Former OpenAI policy chief Miles Brundage weighed in on the recent OpenAI security post, saying nobody informed denies the company has good security people — the real concern is culture and decision-making processes. Echoing Sean Heelan, he argues it's not just the "fucking sandbox": leadership tolerates some non-zero number of incidents, each one externalizing costs instead of internalizing them.

Original post →

More from Companies & People

Companies & People channel →