[OpenAI automated vulnerability link discovered 1,724 Linux kernel vulnerabilities] On November 5th, the OpenAI automated vulnerability research link identified a total of 2,308 Linux kernel vulnerability candidate samples, confirmed 1,724 vulnerabilities through dynamic reproduction, and generated 948 local privilege escalation (LPE) exploits. This type of vulnerability allows unprivileged users to obtain the highest root privileges. The link is divided into five stages: vulnerability discovery, dynamic verification, exploitability assessment, patch generation, and manual review. The entire process of the second scan cost approximately US$1.21 million in API points and produced 948 independent local privilege escalation vulnerabilities. If estimated based on bug bounty standards, the value of these 948 exploits ranges from US$18.9 million to US$66.3 million, which is 16 to 55 times the link cost. The agent uses 10 new vulnerability exploitation techniques, which have been verified as new methods by third-party kernel vulnerability experts. OpenAI is submitting vulnerability information and corresponding patches to Linux kernel maintainers.