Emerging conventions in GenAI disclosure: how applied linguistics scholars disclose AI use

Abstract

Abstract This study examines how applied linguistics authors report the use of generative AI (GenAI) in published research, addressing the ongoing ambiguity around acceptable practices, required disclosure, and informational sufficiency. We conducted a systematic content analysis of 3,384 research articles published between September 2024 and August 2025 in 84 SSCI-indexed applied linguistics journals, focusing on the formal disclosure sections (acknowledgements, declarations and additional information). Only 4.2 % of the articles included any AI-related disclosure: 2.8 % declared use and 1.4 % declared non-use, with reporting concentrated in a minority of journals, indicating policy-driven effects. When present, disclosures most commonly appeared as standalone statements, naming specific tools (especially ChatGPT and Grammarly) and describing language-focused purposes (editing, clarity and proofreading). Human accountability clauses were frequent, asserting author review and responsibility; explicit statements of non-use were brief and standardized in some venues. Limited reporting of data-related uses (e.g., data translation and instrument design) were found. We propose proportionate, structured templates and placement guidance to enhance transparency, comparability, and GenAI literacy for research article writing in applied linguistics.

Ähnliche Arbeiten