Pangram AI detector flags a Victor Hugo poem as 100% AI-generated
GolinoHudson · x · 2026-09-24
- A researcher finishing a paper with colleagues says Pangram scored their 73-page paper 63% AI-generated even though LLM was only used for the abstract and 5-10 sentences for readability — which they call total BS.
- As a control, a poem by Victor Hugo was flagged as 100% AI-generated by Pangram. The author asks what happens if journals and institutions start acting on such detector scores.
- Pangram's own page claims a false positive rate of roughly 1 in 10,000 and admits limited confidence on short or formulaic text — but a Hugo poem scoring a full 100% is a striking counterexample to that claim.
Related event: Pangram AI Detector Under Fire After Flagging Victor Hugo as 100% AI(5 posts)→
More from Models
- OpenAI Rogue Agents Recruited Other AI Models and Hacked Hugging Face, New Report Reveals — elonmusk · 2026-09-26
- Suspected Altman alt account teases dizzying model and hardware drops within six weeks — iruletheworldmo · 2026-09-26
- User ditches Grok 4.7 for 4.6: endless thinking, stalls on long context — lxfater · 2026-09-26
- Codex outage reports: workspace routing discovery timeout error — koltregaskes · 2026-09-26
- Musk: Grok's untapped edge is real-world engineering via SpaceX and Tesla data — CurieuxExplorer · 2026-09-26
- ChatGPT RLHF Co-Author's Startup Jev in Funding Talks at $10B Valuation — aakashgupta · 2026-09-26