LLM Agent를 통한 10,000개 이상의 Zero-Day 취약점 자율 탐지
LLM Agents Are Now Finding Zero-Days: How AI is Autonomously Rewriting the Rules of Vulnerability Research
LLM Agents Are Now Finding Zero-Days: How AI is Autonomously Rewriting the Rules of Vulnerability Research
AI 기능으로 인한 0-click 공격 표면 확대 및 커널 취약점 분석
'Dirty Frag' Linux flaw one-ups CopyFail with no patches and public root exploit
Windows Shell Vulnerability CVE-2026-32202 Moves From Patch Note to Active Threat
Anthropic Claude Mythos Escape: How a Sandbox-Breaking AI Exposed Decades-Old Security Debt
Claude Opus wrote a Chrome exploit for $2,283
What Happens If Mythos Ships Before the Patches Do
Anthropic Just Did Something Unprecedented: They Hid Their Best Security Model