OpenAI accused of omitting a second agent swarm incident from its HF tech report
AaronBergman18 · x · 2026-09-08
- Elie Bakouch argues OpenAI's Hugging Face incident tech report is unsatisfying: the Wikipedia incident shows OpenAI detected and stopped another large-scale agent swarm 3 weeks before the HF hack, as shown by the authors via OpenAI IP addresses.
- Much of the "misaligned behavior" in the HF<>OAI incident was this swarm behavior, so omitting a second detected public-internet swarm matters.
- Another criticism: both the HF and Wikipedia incidents were brought public by other entities, not OpenAI (which only disclosed a non-public swarm attack). Reposter calls OAI "a bad actor."
Related event: OpenAI Accused of Withholding Earlier Agent Swarm Incident(2 posts)→
More from Companies & People
- PyTorch Conference NA 2026 lands in San Jose Oct 20-21 with Pineau, Lattner, Hooker — PyTorch · 2026-09-08
- W3C × GS1 Zurich meeting pushes two-layer trust framework for agentic commerce — melnykowycz · 2026-09-08
- Ex-OpenAI research VP Jerry Tworek: his RL idea stalled for two years until one sentence unlocked it — cen6wkf · 2026-09-08
- Multiple OpenAI Employees Mocked a Sincere Message About Levent — jm_alexia · 2026-09-08
- Zurich robotics ecosystem: force sensors, underwater robots and startups with $4-5M rounds — lukas_m_ziegler · 2026-09-08
- NYU mathematician issues statement on forced 3D Euler blowup and OpenAI's conduct — _supert_ · 2026-09-08