Abstract With the rapid evolution of artificial intelligence (AI) technologies, their potential to support language learning has captured increasing scholarly attention. This systematic review first maps 105 studies over the past three decades on AI-assisted language learning across all levels of education and then offers an in-depth analysis of 71 studies in higher education (HE) contexts to assess the effectiveness of AI-powered tools. Among the studies in HE settings, 51 reported linguistic outcomes, with 92 % demonstrating positive results. Additionally, 41 studies examined non-linguistic outcomes, revealing consistent findings of enhanced motivation and willingness to communicate, but varied findings on learning attitudes, anxiety, and interest. Despite these positive results, the medium risk of bias across studies due to methodological limitations, such as over-reliance on self-report data, suggested that the findings may not be taken as robust evidence for the effectiveness of AI-assisted language learning. The review provides the following theoretical and practical implications: (1) researchers should adopt more rigorous study designs; (2) learners and teachers should keep utilising AI tools to facilitate personalised and interactive language learning; (3) developers should address the identified problems with AI-assisted language learning tools and promote further technological innovations.
Yining Han (Sat,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: