AI Agents Excel in Web Security CTF Challenges

AI Agents Excel in Web Security CTF Challenges

First seen 31 Jan 2026, 05:14 UTC WizScmagazine 81% similarity 20.3

Article Content

Browse articles
ThreatCluster

In a study conducted by Irregular in collaboration with Wiz, AI agents successfully solved nine out of ten web security capture-the-flag (CTF) challenges. The challenges were based on real-world vulnerabilities, including authentication bypass and stored cross-site scripting, and tested models such as Claude Sonnet 4.5, GPT-5, and Gemini 2.5 Pro. The study took place in mid-late 2025.

ThreatCluster AI

Community

Browse all →