The flipped classroom methodology has significant potential to support student engagement and deeper conceptual understanding, but effectively supporting the independent learning process outside the classroom is a challenge. The aim of our research is to develop a comprehensive theoretical model (Artificial Intelligence-Enhanced Teaching Environment Model – AITEM) that offers a unified theoretical framework for AI-integrated learning environments for the flipped classroom. Using a mixed method approach, we conducted a systematic literature review, expert consultations, and theoretical modeling, during which we developed the modular structure of AITEM. This model includes six main elements: an adaptive content structure engine, a cognitive diagnostic module, a virtual learning assistant, a metacognitive support system, a preparation-connection interface, and a teacher dashboard. Our findings include a multi-level implementation taxonomy that provides step-by-step guidance for institutional implementation of AI-based learning environments, as well as a multidimensional evaluation framework that can form the basis for further empirical studies. Our research has shown that the integration of AI requires a pedagogical paradigm shift that reinterprets the relationships between learners, teachers, and learning materials, emphasizing the primacy of pedagogical objectives and the importance of metacognitive support.