Artificial Intelligence for Language Learning: A Systematic Review of its Design, Theoretical Foundations, Implementation, and Impact

ABSTRACT Although the use of AI technologies (e.g., chatbots and automated writing evaluations (AWE)) has gained considerable attention in language learning fields in recent years, how AI technologies have been designed and implemented in language learning education, as well as their effectiveness, is understudied. The current study thus seeks to provide a systematic review of the empirical evidence on the research status and impact of AI technologies in language learning from 2014 to 2023. A total of 30 eligible studies that adopted a multiple‐group experimental design or one‐group pre‐test and post‐test design were selected and systematically analyzed, revealing the following key findings: (1) university students were the most frequently selected samples in the current relevant literature; (2) speaking and writing were the most common target language domains in the current relevant literature; (3) most of the studies utilized AI technologies to facilitate the learning of English as a second or foreign language; (4) the role of AI technologies in language education focused on intelligent tutoring system and intelligent assessment and feedback while less attention was paid to profiling and prediction or adaptive learning systems and personalization; (5) less work had been conducted to orchestrate peer‐to‐peer interactions with the help of AI technologies; (6) few studies designed the use of AI technologies rooted in educational theories; and (7) although the included studies show AI technologies benefit some student language learning outcomes, their effectiveness remain inconclusive given shortcomings in methodological rigor (e.g., the lack of randomized controlled trial). The study concludes by discussing the implications for future research directions and practice.