2026/07/01 by Rubén Comas Forgas, Cinta Gallent Torres, Carmen Touza Garma +1 · 1 voice
Medicine · Social Sciences · #Artificial Intelligence in Healthcare and Education #Computational and Text Analysis Methods #Ethics and Social Impacts of AI
paper · doi:10.1002/leap.2067
openalex publication_date 2026/07/01 · openalex created_date 2026/07/17 · openalex updated_date 2026/07/17
ABSTRACT The widespread availability of generative AI tools is reshaping scholarly discourse at a linguistic level. This study investigates the diffusion of AI‐associated vocabulary in abstracts indexed in Scopus, focusing on post‐2022 patterns that emerged after the public release of tools such as ChatGPT. Using a combined bibliometric and content‐analysis approach, we identified 17 lexical items with unusually steep growth trajectories between 2022 and 2024. These terms were analysed across disciplines, countries, and journals as probabilistic markers of an AI‐era linguistic shift rather than as diagnostic proof of direct AI use in individual articles. The results show a marked post‐2022 lexical acceleration consistent with machine‐mediated or AI‐influenced writing environments, although the underlying causes are likely multifactorial and may include direct use, indirect imitation, editorial normalization, and broader stylistic drift. The gap between explicit AI disclosures and the observed diffusion of these markers is therefore interpreted as an illustrative reporting asymmetry rather than a definitive measure of undisclosed AI use. Overall, the findings point to a substantial transformation in academic expression and call for clearer disclosure taxonomies, stronger editorial guidance, and more nuanced debate about authorship, integrity, and linguistic heterogeneity in scholarly communication.