AI-Powered Writing Assistance Tools in the Foreign Language Classroom: Unveiling ESP Teachers’ Perspectives across Disciplines

Abstract

This mixed-methods study examines Vietnamese English for Specific Purposes (ESP) teachers’ perceptions of AI-powered writing assistance tools and explores how these perceptions differ across academic disciplines. Data were collected from 40 ESP teachers through a questionnaire and semi-structured interviews with 15 participants. Descriptive statistics and thematic analysis were employed to investigate teachers’ attitudes, perceived benefits and challenges, and contextual factors influencing AI integration. Quantitative findings indicate strong agreement on the pedagogical value of AI tools in promoting learner autonomy (M = 4.00, SD = 0.96), enhancing engagement (M = 3.78, SD = 0.91), and supporting collaborative learning (M = 3.82, SD = 0.89). However, teachers expressed significant concerns regarding the accuracy of AI-generated language (M = 4.82, SD = 0.90), students’ overreliance, and difficulties aligning AI tools with existing curricula. Qualitative data revealed notable disciplinary differences: teachers in hard sciences highlighted limitations in domain-specific terminology and precision, whereas those in soft sciences perceived stronger benefits for general academic writing and motivation. The findings underscore the need for discipline-sensitive implementation, clearer institutional guidelines, and targeted professional development to support responsible and effective AI integration in ESP contexts. This study contributes to the emerging literature on AI in language education and offers practical implications for policymakers, curriculum designers, and educators seeking to optimize AI-assisted writing instruction.

Ähnliche Arbeiten