Scanning 24K Pre-LLM Dissertations Shows Near-Zero AI False Positives
birchlse · x · 2026-08-09
A researcher recently conducted an empirical test on the false-positive rate of AI text detectors. By scanning almost 24,000 dissertations written before the advent of LLMs using tools like Pangram, not a single one was falsely flagged as AI-generated.
This indicates that modern AI detection tools are incredibly effective at minimizing false positives, accurately identifying authentic human academic writing without misclassification.
More from Research
- Simplex Diffusion Models author Justin Deschenaux to present his paper Monday — LucaAmb · 2026-10-16
- Tool auto-audits paper claims with pass/fail evals — 4 of 7 metrics reproduce, 1 fails — jwt0625 · 2026-10-16
- Hironobu Fujiyoshi publishes deck: from AI agents to Agentic Robotics — HirokatuKataoka · 2026-10-16
- Is math just a language? A debate on why LLMs are good at mathematics — LucaAmb · 2026-10-16
- Alibaba open-sources OpenGCUA, a full-stack generalist computer-use agent; +19.6 points on OSWorld-v2 across 8 models — Tongyi-MAI · 2026-10-16
- CoDA co-evolves simulation and robot policies via shared failure memory, lifting OOD success from 5.3% to 27.6% — PLAN-Lab · 2026-10-16