From Automated Feedback to Human-AI Collaboration: Mechanisms and Pathways of AI Integration in Foreign Language Education

Abstract

Artificial intelligence (AI) is continuously transforming knowledge organization, teaching activity structure, and learning support mechanisms in foreign language education. Empirical studies in the past five years have shown that AI integration in foreign language education is not a linear process of single-technology application in classrooms, but an evolving trajectory deepening from automated feedback to intelligent interaction and further to human-AI collaboration. Early research mainly focused on automated writing evaluation, automatic speech recognition, machine translation, and basic chatbots, verifying their roles in reducing feedback delay, expanding practice opportunities, and alleviating teachers’ repetitive workload. With the integration of large language models into teaching contexts, research has gradually shifted to higher-level issues such as writing ideation, oral interaction, learning engagement, self-regulation, and human-AI collaboration. Based on a review of representative empirical studies in the past five years, this paper analyzes the evolution, core application fields, main mechanisms, practical limitations, and future directions of AI integration in foreign language education. The study argues that the educational effectiveness of AI is not directly determined by technological advancement, but by the combined effects of task design, feedback arrangement, teacher intervention, learner literacy, and contextual adaptation. Future research should shift from verifying whether tools are effective to explaining how human-AI collaboration works effectively, and construct an interpretable and generalizable research framework in more authentic classroom ecologies, longer time spans, and a wider range of languages.