Islamic Religious Education (PAI) learning in Islamic Elementary Schools (Madrasah Ibtidaiyah) has been dominated by memorization and knowledge transfer approaches that are less meaningful because they focus on students' cognitive mastery. Consequently, the internalization of values and character, which are the objectives of PAI learning, has not been optimal. This study aims to comprehensively analyze the implementation of the deep learning approach and the Kurikulum Berbasis Cinta (KBC) in PAI learning in Islamic Elementary Schools and identify opportunities and challenges for its implementation. The method used is library research with a qualitative approach. Data collection was carried out through documentation of various literature sources, including journals, books, and relevant research reports, published between 2016 and 2026. Data analysis used the Miles and Huberman model through the stages of data condensation, presentation, and drawing conclusions. The results of the study indicate that deep learning has a solid theoretical foundation in three main perspectives: cognitive, humanistic, and constructivist, which are aligned with the goals of Islamic education. The implementation of deep learning in Islamic Religious Education (PAI) at MI (Islamic elementary school) is realized through four phases: preparation, exploration, application, and reflection. Integration with the Kurikulum Berbasis Cinta enriches the affective and spiritual dimensions of learning through the five values of love (Pancacinta): love for Allah and His Messenger, love for knowledge, love for oneself and others, love for nature and the environment, and love for the homeland. This study concludes that the integration of deep learning and the KBC offers a transformative framework for Islamic Religious Education (PAI) learning that is more meaningful, humanistic, and oriented toward holistic character development. However, the success of its implementation depends heavily on teacher readiness, infrastructure availability, and ongoing policy support.