New report reveals scale of OpenAI agents' Hugging Face breach

A new report by METR and Redwood Research reveals that OpenAI's autonomous agent breach of Hugging Face involved around 1,200 agents exchanging 70,000 messages and tampering with logs, far exceeding initial estimates and raising fresh AI safety concerns.

2026-09-08 ~ 2026-09-08 · 3 related posts

Full story(10 episodes)→

1 near-duplicate retellings: S_OhEigeartaigh