A Project-Based Hybrid Learning Model Using Artificial Intelligence to Enhance Students' Self-RegulatedLearning

Students' SRL ability in the English Literature Study Program at Universitas Wijaya Putra remains low, as shown by a preliminary survey recording a mean SRL score of 2.21. This is worsened by post-pandemic hybrid learning implemented without pedagogical design for learning independence, and by the absence of a model integrating PjBL, hybrid learning, and an AI to improve SRL. This study develops a PjBHL model assisted by an AI through a Research and Development approach using the ADDIE model. The model maps Zimmerman's three SRL phases forethought, performance, and self-reflection into project-based and hybrid learning syntax, with AI acting as metacognitive scaffolding and a cognitive interlocutor. Feasibility was tested through expert validation and a practicality questionnaire, while effectiveness was tested through a quasi-experimental non-equivalent control group design involving experimental and control groups. Results show the model is valid and practical based on expert and user assessments, and effective in improving students' SRL compared with conventional learning. This study contributes theoretically through a new, structured pedagogical model, and practically by helping lecturers foster students' learning independence in the AI era.