The application of artificial intelligence (AI) in English as a Foreign Language (EFL) teaching has raised new questions about its effectiveness compared with established models, such as the flipped classroom. In the current study, the relative effect of AI-supported learning compared with flipped instruction on Iranian EFL learners’ self-monitoring ability was examined. Quasi-experimental design was applied to 120 upper-intermediate students who were assigned to control, flipped classroom, and AI-supported learning groups. The control group implemented traditional instruction for ten weeks, whereas the experimental groups were subjected to targeted interventions. Data were collected through standardized self-monitoring questionnaires and written assignments. Analysis of covariance (ANCOVA) revealed that the flipped as well as the AI-facilitated groups were significantly higher than the control group in the acquisition of self-monitoring (p < .01, partial η² = .18). Furthermore, the AI-supported group showed more substantial gains in monitoring during the reflection stage and error detection compared to the flipped group. These findings suggest that while flipped classes promote autonomy under teacher guidance, AI systems provide adaptive scaffolding and individualized feedback to enable learners’ strategic self-regulation ability. The study contributes to CALL and self-regulation research by demonstrating how AI can be leveraged to assist human teaching, making theoretical contributions to learner autonomy models and empirical contributions to the adoption of technology-enhanced practices in EFL teaching.