A Configurational Exploration on the Promise and Perils of Generative AI in Project-Based Language Learning

This study employs fuzzy-set Qualitative Comparative Analysis to investigate how generative artificial intelligence (GAI) integration affects English as a Foreign Language development in project-based learning (PjBL) contexts. This study examined configurations of learner autonomy, collaborative learning, GAI-supported language learning, and GAI-supported project-based learning among 43 undergraduate students engaged in a 12-week intervention. Results identified four pathways to enhanced language development: autonomy with GAI language support, synergistic integration of autonomy and collaboration with GAI in projects, comprehensive GAI integration with collaborative practices, and an individualistic pathway combining autonomy with GAI tools. Necessity analysis revealed that absence of GAI-supported language learning consistently predicted failure. These findings demonstrate the configurational nature of effective GAI integration in language education and provide guidance for implementing these tools across diverse instructional settings.