Mythos 모델의 FOSS 23,019개 취약점 발견 및 보안 거버넌스 수립
Anthropic to release Mythos-class models to the public
Anthropic to release Mythos-class models to the public
AI agents show they can create exploits, not just find vulns
LLM 기반 익스플로잇 가속화로 인한 Coordinated Disclosure 체계 붕괴와 방어 전략의 전환
AI Is Breaking Two Vulnerability Cultures — And Vibe Coders Are About to Get Caught in the Middle
LLM 기반 리버스 엔지니어링을 통한 폐쇄형 API 제어 및 취약점 분석
Anthropic Releases Claude Mythos Preview with Cybersecurity Capabilities but Withholds Public Access
Has Mythos just broken the deal that kept the internet safe?
Project Glasswing: Securing critical software for the AI era