Initial Indications of Generative AI Influence in Linguistics Research Abstracts: Evidence From a Large‐Scale Lexical Frequency Analysis Between 2018 and 2025

Generative AI and large language models (LLMs), such as ChatGPT, have transformed many working practices, including scientific writing. Previous large‐scale studies have identified shifts in the use of certain stylistic markers, such as the word delve , in disciplines like biomedicine, engineering and computer science, yet comparable evidence from the humanities remains scarce. Using lexical frequency analysis, we examined the prevalence of 28 AI‐associated stylistic markers across 129,338 published abstracts from 592 linguistics journals between 2018 and 2025. Counterfactual modelling suggested that approximately 16% of abstracts published in 2025 exhibited frequencies of AI‐associated stylistic markers above levels expected from pre‐LLM trends (95% CI = 14.46%–17.44%). Furthermore, higher‐prestige journals exhibited slightly greater AI‐associated marker word frequency. Country‐level differences indicated particularly higher AI‐word usage in abstracts from Asia and the Middle East. While lexical frequencies provide only indirect evidence of LLM use, the findings suggest that linguistic patterns previously associated with LLM‐assisted writing have become increasingly prevalent in linguistics publications.