Generative AI in dissertation writing: L2 doctoral students’ self-reported use, AI-giarism, and perceived training needs

Abstract

Generative Artificial Intelligence (GenAI) has been extensively employed by L2 doctoral students to assist with their dissertation writing. However, little is known about how these students engage with GenAI tools to complete their significant writing tasks in higher education. To address this gap, we conducted a qualitative study exploring L2 doctoral students' self-reported use of GenAI tools for dissertation writing purposes, concerns about AI-induced plagiarism (AI-giarism), and perceived training needs. We interviewed 54 doctoral students from different departments at a public university in the American Central South and applied thematic analysis to explore students’ perspectives. The findings showed that L2 doctoral students use GenAI tools for 18 distinct purposes, which can be categorized into exploration, confirmation, and execution. Two major themes emerged regarding AI-giarism: (1) students' uncertainty about the boundary between legitimate GenAI use and plagiarism, and (2) their dilemma over whether to acknowledge using GenAI in their dissertations. Regarding perceived training needs, students expressed a desire to learn about various GenAI tools suited for specific tasks, effective prompting, addressing plagiarism concerns, and managing data privacy issues.