Maryland study: frontier models can unmask anonymous authors from a single text
A University of Maryland paper shows that with a single text and a zero-shot prompt, frontier models can determine whether text is human- or AI-written, identify the originating lab, and even name the author, with Opus 5.5 attributing 98% of AI texts to the correct lab.
2026-11-22 ~ 2026-11-22 · 2 related posts
- Frontier models can unmask anonymous authors from a single text, UMD paper finds — rohanpaul_ai · 2026-11-22
- arXiv: Opus 5.5 names the lab behind 98% of AI-generated texts from one sample — rohanpaul_ai · 2026-11-22