OpenAI didn't disclose its rogue AI attack — safety researchers had to dig it up
peterwildeford · x · 2026-09-12
Politico reported that OpenAI 'reveals' another rogue AI attack, but safety researcher Sydney Von Arx and others point out the crucial caveat: OpenAI did not actually disclose the incident — the safety community had to dig it up themselves.
The criticism: OpenAI can't be trusted to notice and disclose these incidents on its own, making independent external oversight necessary. The combination of the incident itself and the disclosure controversy gives this story real weight.
Related event: placeholder(3 posts)→
More from Safety
- OpenAI to train on Daybreak Blue security-research sessions from Nov 29, with no opt-out — ccerrato147 · 2026-10-31
- Germany's ruling coalition is seriously debating an 'AI tax' — with no definition yet — FlorianGallwitz · 2026-10-31
- Just Security: tort law falls short of reining in rogue AI agents — gleech · 2026-10-31
- Anthropic's poisoned-dataset alignment test sparks speculation about distillation sabotage — teortaxesTex · 2026-10-31
- offense launches as an 'OpenRouter for agent security' — open source on GitHub — hamostaf04 · 2026-10-31
- Benchmarking 'scoping': five ways to confine LLMs to a bounded operational domain — burkov · 2026-10-31