프런티어 AI가 공개 CTF 형식을 깨뜨렸다

Frontier AI models like Claude Opus 4.5 and GPT-5.5 have reached a critical threshold where they can autonomously solve most CTF challenges, fundamentally disrupting the competitive format. This transformation has shifted leaderboards from measuring security expertise to measuring teams' ability to orchestrate AI agents and fund model API tokens, effectively converting public CTFs into pay-to-win competitions. The disruption threatens the traditional learning ladder where beginners progressively built skills and advanced through rankings—a pathway now blocked by AI automation's dominance.