Frontier models can unmask authorship from a single text, no tools needed
MohitIyyer · x · 2026-11-20
New research thread from atreydesai: given a single text, frontier models can often tell whether AI wrote it, which lab's AI, or even the specific human behind it—no web search, no tools, no candidate list. MohitIyyer notes authorship attribution is advancing extremely fast, with major deanonymization and impersonation risk.
More from Models
- Professors built 'AI-proof' exam questions — GPT-5.4 still scored 83% — alex_verem · 2026-11-20
- User tests Opus 5.5 via Claude Pro: after 24h it beats ChatGPT subs on value — mariofilhoml · 2026-11-20
- Google AI declares a border configuration 'mathematically impossible' and proves it — wrongly — conitzer · 2026-11-20
- 13 AI models vs. real policymakers on policy questions: 10 models beat the humans — ChenhaoTan · 2026-11-20
- Contralabs launches Frontier Design Benchmark with expert-correlated design rubrics — floguo · 2026-11-20
- Redditor slams Google AI search: six months of updates made it 'cheaper, weaker, gaslighting' — GlitterHeart-Bliss · 2026-11-20