DARPA's post-AIxCC bounty program led Xint.io to discover CVE-2026-31789, a critical heap buffer overflow vulnerability in OpenSSL that could enable code execution. Xint's AI-powered scan analyzed ove…
A self-orchestrating team of LLM-driven vulnerability-hunting agents has discovered over 20 CVEs in recent months, including CVE-2026-31432 and CVE-2026-31433, two remote, unauthenticated out-of-bound…
Xint Code is a new AI-powered code analysis tool that autonomously discovered critical zero-day remote code execution (RCE) vulnerabilities in Redis, PostgreSQL, and MariaDB without any human interven…
Theori's open-sourced Cyber Reasoning System (CRS) called RoboDuck, developed for DARPA's AI Cyber Challenge (AIxCC). The system uniquely uses an "LLM-first" approach to find bugs and generate Proofs …
Strategies for building effective LLM agents for the AI Cyber Challenge (AIxCC), where teams create autonomous Cyber Reasoning Systems (CRS) to find and patch vulnerabilities. The authors advocate for…