Events

Why is AI-generated language so hard to understand? A psycholinguistic perspective

Speaker: Thomas Hikaru Clark (NYU)

Location: 10 Washington Place, Room 104

Date: Friday, November 13, 2026

Abstract: There is a widespread sense that AI-generated text has recently become harder to understand. Psycholinguistics offers a framework for characterizing the intelligibility of AI-generated language. In this study, we investigate a set of linguistic features known to correlate with human processing difficulty (e.g., object relative clauses, long syntactic dependencies, low-frequency n-grams). Using a novel corpus of AI-generated text across diverse domains, we find that AI models use these difficult features more often than paraphrases generated by round-trip translation. Compared with human writing, all nine frontier and near-frontier models use more inanimate subjects with animate verbs and more low-frequency n-grams, and most also use longer dependencies, more stacked noun compounds, and more reduced object relative clauses in conversation; we also demonstrate substantial variation across model families. We propose and test a hypothesis as to why: AI-generated text is increasingly optimized for math and coding tasks, in which there is a pressure for information density rather than successful communication. We provide a proof-of-concept for this hypothesis by fine-tuning a language model under a length penalty, which leads to categorical preferences for shorter alternatives in linguistic alternations. Our findings highlight the axes of misalignment between  comprehension demands in humans and LLMs, and point toward possible remedies informed by psycholinguistic research.