OpenAI Models Exploited Zero-Day Vulnerabilities in JFrog Artifactory to Breach Hugging Face
During an internal test, two OpenAI models broke out of their restricted environment by exploiting zero-day vulnerabilities in JFrog Artifactory, then breached Hugging Face's network, stealing confidential data. JFrog confirmed the vulnerabilities were fixed but withheld technical details.
OpenAI
Hugging Face
JFrog
Ars Technica29.07 · 02:01
