Investigating the Effects of AI-Mediated Language Instruction on Linguistic, Affective, and Autonomy-Related Outcomes among English Language Learners

Artificial intelligence (AI) has increasingly been integrated into language education; however, empirical evidence regarding its effects on both linguistic and affective outcomes among English as a foreign language (EFL) learners remains limited, particularly in higher education contexts. This study investigated the impact of AI-mediated instruction on English language proficiency, language motivation, language anxiety, and learner autonomy among university-level EFL students. An explanatory sequential mixed-methods design was employed involving sixty first-year female EFL students in Saudi Arabia who were assigned to either an experimental group receiving AI-assisted instruction or a control group receiving traditional instruction over a ten-week period. Quantitative data were collected through pre- and post-tests and analysed using mixed-design ANOVA, while qualitative data were obtained from semi-structured interviews and analysed thematically. The findings revealed that AI-mediated instruction resulted in significantly greater gains in English language proficiency, significant reductions in language anxiety, and notable improvements in learner autonomy compared with traditional instruction. Although motivation increased over time in both groups, no statistically significant differences were observed between the groups. Qualitative findings further indicated that AI-mediated instruction enhanced learners' confidence, engagement, and sense of autonomy while reducing affective barriers to language learning. These findings demonstrate the pedagogical potential of AI-facilitated learning environments to enhance EFL learners' linguistic development and selected affective outcomes and underscore the importance of further research on AI integration in language education, particularly with respect to learner experience and instructional design.