Anthropic’s AI model Claude cracks top ranks in student hacking contests

Anthropic’s AI model Claude has been quietly outperforming human competitors in student hacking contests, revealing how far artificial intelligence has advanced in offensive security.

Over the past year, Claude has quietly emerged as a top contender in PicoCTF, the flagship capture-the-flag cybersecurity contest run by Carnegie Mellon University, Axios reported on Tuesday (August 5). The AI breezed through beginner-level challenges like decryption and reverse engineering — routines typically designed to test the skills of high school and college…