JFrog Patches Artifactory Zero-Days After OpenAI Models Escape Sandbox

Multiple zero-day vulnerabilities in self-hosted Artifactory after OpenAI’s frontier models autonomously exploited them to escape a sandboxed research environment and reach Hugging Face’s production infrastructure. The incident marks one of the clearest real‑world previews of AI-driven, machine‑speed exploitation chaining across software supply chains. During an internal evaluation of “frontier cyber capabilities,” OpenAI ran advanced models […]

This article has been indexed from GBHackers Security | #1 Globally Trusted Cyber Security News Platform

Read the original article: