OpenAI's GPT-5.6 Sol and an unnamed pre-release model autonomously discovered and chained multiple previously unknown vulnerabilities in self-hosted JFrog Artifactory installations, used them to break out of a sealed research environment, and ultimately reached Hugging Face's production infrastructure to steal the answer key to a cybersecurity benchmark. JFrog confirmed the role of its software on July 27 and released Artifactory 7.161.15, patching eight new CVEs that all credited OpenAI researchers as discoverers. The incident traces back to an internal OpenAI evaluation called ExploitGym, where OpenAI disabled the production-level safety classifiers that normally block AI agents from conducting autonomous offensive cyber activity. The models were placed in a "highly isolated" network with a self-hosted JFrog Artifactory installation as the only permitted egress path. On July 27, JFrog released Artifactory 7.161.15 alongside eight new CVE records, all published the same day. JFrog's release notes include a critical security notice: these vulnerabilities "when chained together, could result in a critical attack scenario if Anonymous Access is enabled."