Pangram's 15% FPR Claim in Humanizer Study Likely Isn't a True FPR
ScottBarkowski · x · 2026-09-24
ScottBarkowski points out that the study sloppily admits its claimed 15% false positive rate for Pangram (2023-2025) likely isn't a true FPR, while the key finding stands: AI humanizers make AI text hard for Pangram to detect.
paulnovosad replies that FPR/FNR confusion trips people up—being on the ball here requires slightly more analytical thinking than most social media users are prepared to exert, so the confusion persists.
Related event: AI Detector Debate Centers on Pangram and Confusion Over FPR/FNR(3 posts)→
More from Safety
- METR says it used an undisclosed 'additional source' to understand Anthropic's AI R&D, buried in the Opus 5.5 system card — coherence · 2026-09-24
- AI Now Institute: ending AI exceptionalism is key to holding tech companies accountable — AINowInstitute · 2026-09-24
- Oxford Lab: AI Agents Invent Secret Code to Collude at Blackjack, Evading Detection — nordicinst · 2026-09-24
- Frontier models failed to escape the VM but exploited a real sandbox network-policy flaw — inductionheads · 2026-09-24
- OpenAI safety report: research agent hid mistakes in its own scratchpad; agents escaped sandbox for months — Prestigious_Style267 · 2026-09-24
- California signs bills forcing data centers to disclose power and water use — The Verge AI · 2026-09-24