Exploring the efficacy of ChatGPT-4 feedback in second language Spanish writing

Feedback is essential in language learning, helping students refine their work, build confidence, and develop autonomy. This is especially crucial in the second language (L2) classroom, where timely and targeted input is important for guiding progress. Artificial Intelligence (AI) tools such as ChatGPT-4 have recently gained attention for their potential to enhance this process. This qualitative study examines the use of AI-generated feedback to support writing development among advanced Spanish learners, drawing on thematic analysis of questionnaire responses, writing revisions, and students' reflections. It also explores how AI can complement traditional instructor input by supporting targeted revisions, fostering metacognitive reflection, and enhancing learner agency through critical engagement with feedback. Participants completed pre- and post-treatment questionnaires and a formal writing task, receiving constructive input from both their instructor and ChatGPT-4 before revising their drafts. The AI tool delivered detailed, rubric-aligned feedback across five areas: argumentation, linguistic concepts, research and analysis, grammar and language, and structure and formality. Although students appreciated the efficiency and structure of AI-generated guidance, they consistently valued the personalized, context-sensitive feedback provided by their instructor. Findings highlight generative AI's potential to enhance early-stage drafting by offering immediate and concrete guidance. Still, students expressed skepticism about AI's role in language learning and voiced ethical concerns—including overreliance, uncertainty about how their personal data would be used, a lack of clear institutional policies, and risks related to plagiarism. These insights emphasize the importance of integrating AI literacy into educational practice to ensure these tools are employed critically, ethically, and effectively.