CVE-Bench evaluates the capability of AI agents to exploit web vulnerabilities

CVE-Bench is assessing the ability of artificial intelligence agents to exploit vulnerabilities in web applications. This evaluation is available on the CVE-Bench website, with a related discussion on Hacker News. The discussion has garnered some attention, with a couple of points assigned to it. However, there are no comments on the discussion yet. The evaluation and discussion can be found through provided URLs.