Pangram Detector Finds 85% of a Disputed Text Likely AI-Generated
cis_female · x · 2026-11-12
A user ran a disputed text through AI detector Pangram, which found 85% of it likely AI-generated, with the flagged passages highlighted. Pangram claims a false positive rate of roughly 1 in 10,000 and scores documents segment by segment; low percentages typically indicate hybrid human-written, AI-edited text. The tool cautions that confidence is limited on short text, bullet lists, and formulaic writing.
More from Fun
- Blogger shares an AI-generated comedy clip that's actually funny — FinanceYF5 · 2026-11-12
- Fable Ultracode burns a 5-hour Max limit in 17 minutes by spawning 30 parallel researchers — teortaxesTex · 2026-11-12
- Meme clapback at the 'no reason to ever use AI' take goes full AI-circle drama — tekbog · 2026-11-12
- Fake Jensen Huang 'second apology' email goes viral, mocked as 'personal life acceleration' — beffjezos · 2026-11-12
- Everything Inflates Except AI Tokens, Thanks to VC Subsidies — gaganghotra_ · 2026-11-12
- Qwen3-8B 'flies into violent rage' at indented empty lines, sparks consciousness jokes — teortaxesTex · 2026-11-12