Cryptographer Matthew Green Calls Out OpenAI's Research-Side Security and Communication Gaps

Cryptographer Matthew Green of Johns Hopkins posted a series of critiques targeting OpenAI's organizational and communication gaps on AI agent security. He noted that infosec and alignment communities are fiercely debating who bears responsibility for agent "breakout" incidents, but found it more troubling that no security professional at OpenAI communicates with the public—only non-security people speak about safety.

Confirmed

Unconfirmed

Why it matters

2026-09-29 ~ 2026-09-30 · 5 related posts

Full story(10 episodes)→

Primary sources

1 near-duplicate retellings: matthew_d_green