Frontier AI has broken the open CTF format

Frontier AI models like Claude Opus 4.5 and GPT-5.5 have fundamentally broken the open CTF (Capture The Flag) competition format by automating the solving of medium and hard difficulty challenges, reducing the competition from a test of security skills to a matter of who can afford to run the most AI agents. The author, a top-ranked CTF competitor, argues that AI's ability to independently reason through and solve challenges—rather than merely assist players—has transformed leaderboards into measures of orchestration and token spending rather than cybersecurity expertise, driving away both experienced players and challenge developers.