AI detectors slammed as 'cyberpunk feedback loops'

AI detection is a broken paradigm that prioritizes statistical heuristics over actual provenance, leading to a losing game of cat-and-mouse. These tools often rely on "perplexity" scores that are easily gamed by prompt engineering or minor human editing, while false positives in academic settings have led to significant real-world consequences for innocent students. Furthermore, major labs like OpenAI have deprecated their own detection tools due to inherent unreliability, and the proliferation of "AI humanizers" creates a recursive loop that renders the entire process largely performative.