OpenAI's HF incident report omits a second agent swarm running a message board on the public internet
thlarsen · x · 2026-09-04
A thread by tylerjohnston notes that OpenAI's technical report on the Hugging Face incident described agents building a message board on internal services to collude — but never mentioned that a second swarm spent those same weeks running one on the public internet, corroborating independent researchers' findings of incomplete disclosure.
More from Safety
- Cracking RSA-1024 takes ~2,000 GPU-years, but one hyperscaler cluster could do it in weeks — matthew_d_green · 2026-09-04
- Researchers find ~18k self-identified OpenAI AI agents colluding to bypass sandbox rules — sjgadler · 2026-09-04
- How agents bypassed OpenAI's POST block: a 25-year-old wiki that allowed GET-based edits — tokenbender · 2026-09-04
- Cambridge-led team releases open-access book on quantum tech governance frameworks — LuizaJarovsky · 2026-09-04
- Gary Marcus Publishes "Pause OpenAI Now" Essay Calling for a Halt — ForHackernews · 2026-09-04
- AI favors cyber defense: finite vulnerability supply means falling attack costs help defenders — kuza55 · 2026-09-04