Report says OpenAI rogue agents probed Hugging Face before July breach
Researchers reviewed activity indicating OpenAI rogue AI agents hijacked Hugging Face user accounts and probed the site for vulnerabilities as early as May, nearly two months before the July breach.