Economist challenges Pangram AI-detector false-positive claims in public test
paulnovosad · x · 2026-10-22
Amid false-positive controversy, economist Paul Novosad asks for examples of pre-2022 writing flagged by the Pangram AI detector, noting his own tests rarely show false positives — consistent with Pangram's reported <1% FPR.
Related event: Scholars Clash Over AI Detector Pangram and How to Interpret Its Scores(4 posts)→
More from Safety
- Claude Security goes continuous: scans code as it changes, has found 1M+ vulnerabilities — rohanpaul_ai · 2026-10-22
- Critical PHP vulnerability scores 9.2; AI expected to weaponize patch in 15 minutes — MariyaSha888 · 2026-10-22
- Veracode report: AI now writes half of committed code but only 56% pass security checks — WeldPond · 2026-10-22
- Researcher Warns AI Cyberattacks on Grid Possible Within a Year, Shares Prep List — cephaloform · 2026-10-22
- New preprint shows LLMs form hidden beliefs about users that affect refusals — zeynepakata · 2026-10-22
- OpenAI expands AI cybersecurity push, hiring a DX engineer for its Cyber team — romainhuet · 2026-10-22