I’ve tried using LLMs to deobfuscate libraires like fingerprintjs-pro to understand what specific heuristics implementation details they use to detect bots. Th

I’ve tried using LLMs to deobfuscate libraires like fingerprintjs-pro to understand what specific heuristics implementation details they use to detect bots. They mostly fail. A human reverse engineer will still do better.