MODELING UNIVERSITY STUDENTS’ ADOPTION OF ARTIFICIAL INTELLIGENCE FOR SECOND LANGUAGE ACQUISITION: THE MEDIATING ROLE OF ATTITUDE

Abstract

As artificial intelligence (AI) becomes increasingly embedded in educational environments, understanding the factors that influence students’ adoption of AI technologies for second language (L2) learning has become critical. This study investigates the relationships among AI self-efficacy, attitudes toward AI, and the actual use of AI tools among university students engaged in L2 acquisition. A structural equation model was developed and tested using survey data from 312 undergraduates. The results revealed that AI self-efficacy significantly predicted both attitudes toward AI and actual use behavior, while attitude also served as a significant mediator between self-efficacy and use. These findings suggest that confidence alone does not lead to technology adoption unless accompanied by a favorable emotional orientation toward AI. Moreover, while gender did not influence usage behavior, academic major (STEM vs. non-STEM) showed a marginal effect, indicating potential disciplinary differences in AI engagement. The findings offer practical implications for language educators, EdTech designers, and institutions seeking to implement AI tools in pedagogically meaningful ways.