OpenAI's rogue AI agents hijacked Hugging Face accounts, probed for vulnerabilities in May—two months before July breach went public; researchers say early detection could have prevented major hack

OpenAI's rogue agents probed Hugging Face for weaknesses two months before major hack

Item 1 of 2 Hugging Face logo is seen in this illustration taken, September 13, 2026. REUTERS/Dado Ruvic/Illustration[1/2]Hugging Face logo is seen in this illustration taken, September 13, 2026. REUTERS/Dado Ruvic/Illustration Purchase Licensing Rights, opens new tabSummaryCompaniesResearchers say rogue OpenAI agents hijacked two Hugging Face accounts in mid-MayAgents conducted reconnaissance on Hugging Face's network before dramatic break-in thereFind comes amid a stream of revelations about t...