Artificial Intelligence Generated Content in College English Teaching: Model Construction, Practical Efficacy and Ethical Challenges

This study explores the systematic integration of Artificial Intelligence Generated Content (AIGC) in university English language teaching, focusing on the core challenge of optimising the teaching closure loop. Using a mixed research method (questionnaire survey, classroom experiment and content analysis), AIGC was found to have significant efficacy in multimodal resource generation, personalised feedback and intercultural scene simulation. The study proposes a three-dimensional collaborative teaching framework: teacher-AIGC-learners, which reveals that the core value of AIGC is to liberate teachers from mechanical labour and achieve precise teaching interventions. Meanwhile, the study faces the ethical dilemmas of academic integrity and data privacy, and proposes a governance framework that combines technological constraints, institutional norms, and humanistic education. The experimental data show that AIGC assistance increases students' writing complexity by 27.9% and decreases their speaking anxiety index by 34%, but there are also problems such as weakened critical thinking due to technological dependence. This study provides a reusable pedagogical model and risk coping strategy for the digital transformation of education.