An AI detector correctly labels four AI essays and two human ones, then the writer jokes that only machines can tell

paul_cal · x · 2026-07-25

A quoted passage describes an experiment with Pangram, an AI-writing detector, claiming it correctly labeled four AI essays as AI-written and two human essays as human-written with high confidence.

The punchline is the irony: the writer says fans can barely tell their writing from Claude’s, but another AI can — so their individuality is “still there,” just detectable by machines rather than humans.

Original post →

More from Fun

Fun channel →