Human researcher Jaeyoung Chung uncovered the Bad Epoll kernel race‑condition (CVE‑2026‑46242) that AI model Mythos missed, demonstrating that even the most advanced AI code‑auditing tools can overlook critical vulnerabilities while human intuition and deep analysis still dominate security research.
Anthropic's Claude Opus 4.8 uncovered a four‑year‑old vulnerability in Zcash's Orchard privacy protocol that lets attackers mint unlimited fake ZEC, leading to a near‑50% price plunge, a rapid emergency fix, and a broader signal that AI‑assisted security audits are becoming a reality in blockchain.
The author demonstrates how a human‑provided prompt combined with DeepSeek v4 Pro and Claude Code can autonomously audit the Linux 6.12 crypto subsystem, locate the CVE‑2026‑31431 “Copy Fail” privilege‑escalation bug, and validate the full exploit chain in four iterative dialogues costing less than three dollars.
The EXPRESS 2025 workshop at ISSTA in Norway will showcase AI‑driven code auditing, present cutting‑edge research on trustworthy software systems, and invite researchers and practitioners to discuss transparency, reliability, and security challenges in modern software engineering.
