Paper: AI Text Is Hard to Read Because LLMs Optimize for Density, Not Communication
rickasaurus · x · 2026-10-26
An arXiv paper by Clark, Martin, and Mahowald investigates why AI-generated text has become harder to understand. Using a novel multi-domain corpus and psycholinguistic features known to correlate with human processing difficulty (object relatives, long syntactic dependencies, low-frequency n-grams), they find all nine frontier and near-frontier models use more inanimate subjects with animate verbs and more low-frequency n-grams than humans, with substantial variation across model families. Their hypothesis: human language optimizes for successful communication, while AI text is increasingly optimized for math, code, and reasoning — favoring information density. Fine-tuning a model under a length penalty reproduces the divergence from human-like language, pointing to axes of misalignment and possible remedies. Commenters quip that soon everyone with an IQ over 120 will be employed to talk to LLMs.
Related event: AI Writing Is Getting Harder to Read, Study Finds(3 posts)→
More from AGI Musings
- New paper: AI agents negotiate badly under uncertainty and break deals on purpose — menhguin · 2026-10-26
- NBER: Only 2.9% of US consumers pay for AI; retention as low as 20%, pricing power at risk — avicgoldfarb · 2026-10-26
- AI researchers guessed robots wouldn't outrun humans until 2034 — it already happened — ben_j_todd · 2026-10-26
- AI folks, stop saying "biology" when you mean drug development — KevinKaichuang · 2026-10-26
- ESR on the Bugpocalypse: locking down frontier models only hands the game to Chinese open weights — colinmcnamara · 2026-10-26
- Anthropic: junior tech hires fell from 27% to 9% as agent hours hit 6x employee hours — thione · 2026-10-26