THE EFFECT OF CHATGPT-ASSISTED FEEDBACK ON EFL STUDENTS' ACADEMIC WRITING SKILLS

Аннотация

The rapid diffusion of large language models into higher education has reopened long-standing debates about the role of feedback in second-language writing development. This article synthesizes the growing body of empirical research on ChatGPT-assisted feedback and its effect on the academic writing skills of English as a Foreign Language (EFL) learners. Drawing on quasi-experimental, mixed-methods, and qualitative studies published between 2023 and 2026, the review examines how ChatGPT-generated feedback influences accuracy, organization, coherence, and learner motivation, and how it compares with teacher-generated and peer feedback. The evidence indicates that structured ChatGPT feedback produces measurable gains in grammatical accuracy, lexical range, and textual organization, and in several studies performs comparably to teacher feedback, while learner engagement depends heavily on feedback literacy and prompt design. The article also considers persistent concerns regarding over-reliance, feedback quality variability, and the erosion of learner autonomy. It concludes with pedagogical recommendations for integrating ChatGPT as a complementary, rather than a substitutive, feedback source within EFL writing instruction, and outlines directions for future longitudinal and cross-proficiency research