AI-powered feedback in second language writing: A systematic scoping review of technological innovations in digital language education

This study systematically reviews empirical research on the use of AI-driven feedback systems in second language (L2) writing instruction. It examines their effectiveness in improving learners’ writing proficiency, with particular attention to grammar, vocabulary, coherence, and argumentation. A systematic review was conducted following PRISMA guidelines. Twenty-four peer-reviewed empirical studies published between 2020 and 2024 were selected based on predefined criteria. A mixed-methods synthesis was used to analyze outcomes related to linguistic proficiency, feedback characteristics, learner engagement, and pedagogical integration. AI-powered feedback systems are consistently effective in enhancing lower-order writing skills—especially grammatical accuracy and lexical diversity—among intermediate-level L2 learners. Studies suggest that combining AI and human feedback yields the most comprehensive instructional benefits. It is suggested that AI tools may foster learner autonomy and reduce teacher workload.