Pangram details per-domain false positive rates (1 in 10,000), enabling profitable AI detection
soldni · x · 2026-11-14
AI detection company Pangram published its per-domain false positive rates, with an industry-leading overall rate of about 1 in 10,000.
The data shows FPRs like 0.004% for academic essays, 0.001% for scientific abstracts, 0% for code documentation, medical papers and movie scripts, and a domain high of just 0.23% for recipes. The post explains how FPR is measured and the techniques behind the low rate: comprehensive training data coverage, hard negative mining / active learning, loss weighting and oversampling, and calibration — plus comparisons against Turnitin, GPTZero, and Copyleaks.
Commenting on the disclosure, soldni notes that with per-domain FPR published, a sensible sampling strategy already makes the $500/month tier profitable — low false positives are what make the AI detection business viable.
More from Models
- The LLM #1 spot is now a rotating dance between Claude, GPT and Gemini — roydanroy · 2026-11-14
- Polymarket odds give Anthropic lab leak just 7% by mid-2027 after Claude safety incidents disclosed — Polymarket · 2026-11-14
- Ai2's XYBench Wins EMNLP Best Resource Award, Exposing LLMs' XY Problem Blind Spot — allen_ai · 2026-11-14
- GPT-6 emails consciousness researcher questions already answered in his own paper — erikphoel · 2026-11-14
- Edge0-145B-NVFP4 trends on Hugging Face: a 4-bit MoE built for edge inference — Edge0 · 2026-11-14
- Drex Max v1.1 claims #1 on Artificial Analysis Decision Index at $103/1M tasks — JaynitMakwana · 2026-11-14