A Preliminary Study on Using Large Language Models in Software Pentesting · HackerPapers discussionshttps://hackerpaper.kaipfister.de/p/2401.17459/Collected HN comments and submissions linking this paper.enMon, 02 Nov 2026 11:16:22 GMTA Preliminary Study on Using Large Language Models in Software Pentesting — mentioned by ycombinator5689https://news.ycombinator.com/item?id=39215851urn:hackerpapers:mention:2401.17459:39215851Thu, 01 Feb 2024 13:48:43 GMT<p>A recent study, &quot;A Preliminary Study on Using Large Language Models in Software Pentesting,&quot; explores the potential of large language models (LLMs) in automating security tasks, specifically in software penetration testing. The research investigates how LLM-based AI agents can be improved over time through interaction with human operators and prompt engineering. Utilizing the OWASP Benchmark Project 1.2, the study compares the performance of these AI agents against traditional static code analyzers like SonarQube, highlighting the viability of LLMs in enhancing software security vulnerability…</p><p><a href="https://hackerpaper.kaipfister.de/p/2401.17459/">Paper and discussion history</a>. A mention is not an endorsement.</p>cs.CR