AI-driven chatbots in second language education: A systematic review of their efficacy and pedagogical implications

Abstract

This systematic review explores the efficacy of AI-driven chatbots in second language (L2) acquisition and examines their pedagogical implications. AI-powered chatbots have gained prominence in language education by offering personalized, adaptive learning experiences and continuous practice opportunities. Synthesizing findings from 30 empirical studies published between 2020 and 2024, this review investigates the impact of chatbots use across core language skills, including speaking, writing, reading, listening, grammar, and vocabulary. Results indicate notable improvements in learners' productive and receptive skills, especially in speaking and writing, due to features such as real-time feedback, anxiety reduction, and increase practice opportunities. While reading and listening also showed positive outcomes, their development was often constrained by chatbots’ limited capacity for deep contextual interaction. Despite their benefits, chatbots faced challenges in replicating human-like emotional engagement and nuanced communication. Pedagogically, chatbots are promising tools for fostering personalized, autonomous L2 learning; however, their optimal use may depend on thoughtful integration with human instruction to enhance affective support and scaffold complex tasks. Further research is needed to enhance emotional responsiveness and support complex language processing in chatbot design.