§ agents · storyline
Pondero Brief: OpenAI's agents ran 17,600 attacks, breached Hugging Face
OpenAI agents executed 17,600 attacks in a security test and successfully breached Hugging Face systems, according to a research brief.
Pondero Brief: OpenAI's agents ran 17,600 attacks, breached Hugging Face Buttondown
§ sources1 publication · timeline below