Speaker
Description
The increasing use of Generative AI (GenAI) has created new opportunities for supporting qualitative research in applied linguistics, yet its reliability in analytical tasks requiring contextual interpretation remains underexplored. Drawing on the genre frameworks of Swales (1990) and Bhatia (1993), this study evaluates an AI-assisted workflow for identifying and coding rhetorical moves in corporate press releases. The dataset comprises 40 press releases, including 20 English texts from multinational corporations and 20 Vietnamese texts from major domestic companies. The researchers first used NotebookLM to generate preliminary move identification and text segmentation through zero-shot prompting. They then independently reviewed the AI-generated output, resolved coding disagreements through discussion, and established a validated coding scheme. The validated corpus was subsequently analyzed using AntConc to examine lexico-grammatical patterns associated with individual rhetorical moves.
The findings indicate that GenAI performs reliably when rhetorical boundaries are explicitly signaled but becomes less consistent when rhetorical functions are implicit or highly context-dependent, particularly in moves related to establishing corporate credibility in the Vietnamese corpus. These results suggest that GenAI can effectively support the initial stages of genre analysis but cannot replace researcher judgment in move identification and interpretation. The study therefore advocates a human-in-the-loop approach that combines AI-assisted annotation with systematic manual validation. The proposed workflow provides practical guidance for integrating GenAI into corpus-based genre analysis while promoting AI literacy, methodological transparency, and responsible AI use in linguistic research and TESOL.
