{"ok":true,"trend":{"id":408651,"platform":"hn","region":"global","key":"every model (incl. jev) we tested inflates security finding severity","title":"Every model (incl. Jev) we tested inflates security finding severity","url":"https://casco.com/blog/jev-cvss-benchmark","first_seen":"2026-09-29T22:14:47.436737Z","last_seen":"2026-09-29T22:14:47.436737Z","last_rank":12,"peak_rank":12,"last_volume":8,"peak_volume":8,"seen_count":1,"score":0.7784375,"category_hint":"football","section":"sport","category":"football","summary":"Security firm Casco reports that every large language model it tested, including its own Jev model, inflated the severity of security findings when scoring vulnerabilities, overstating risk compared to expected CVSS ratings. The company published a benchmark detailing the results, prompting discussion about how far AI-generated severity scores can be trusted in security workflows.","why":"It raises fresh concerns about relying on AI models for vulnerability assessment at a time of rapid AI adoption in security teams.","tone":"neutral","entities":["Casco","Jev","CVSS"],"summarized_at":"2026-09-29T22:16:38.057328Z","meta":{"via":"scan","lang":"en","link":"https://casco.com/blog/jev-cvss-benchmark","term":"nfl","comments":1},"nw":null,"promo":null,"kind":null,"importance":null,"hidden":false,"hide_reason":null,"judged_at":null,"title_en":"Benchmark finds AI models inflate security vulnerability severity","section_name":"Sport","category_name":"Football","timeline":[{"captured_at":"2026-09-29T22:14:47.436737Z","rank":12,"volume":8}],"posts":[{"platform":"hn","url":"https://news.ycombinator.com/item?id=49894699","author":"brene","title":"Every model (incl. Jev) we tested inflates security finding severity","snippet":null,"posted_at":"2026-09-29T15:11:25Z","likes":8}],"elsewhere":[],"window":"7d"}}