研究发现使AI文本难以阅读的六种模式

a study by new york university and the university of texas at austin examined 14 language models, including claude, gpt-5.6, glm-5, kimi k2.5, and smaller open-source models like llama and olmo. researchers analyzed 120 conversations, 45 coding sessions, and 980 essays, comparing them to human-written text. they identified six linguistic patterns that make ai text hard to read, such as unusual word combinations, nested sentences, and metaphors. smaller models like llama and olmo produce text closer to human writing. the study suggests that training rewards information density over clarity, contributing to the issue.