<![CDATA[Artificial Intelligence (AI) has the potential to enhance self-regulated learning (SRL) by offering adaptive feedback and personalized support. However, the development of AI-based Learning Management Systems (LMS) tailored to SRL in secondary mathematics education remains limited, particularly in the Indonesian context. This study addresses the need for a pedagogically grounded system by focusing on the Define phase of the 4D model to identify learner needs and design requirements. A research and development (RD) approach was employed, using the Define phase of the 4D instructional design model. A needs analysis was conducted through a survey of 805 secondary and university students across Indonesia and a limited system trial with 26 teachers and students. Quantitative data were analyzed using descriptive statistics, complemented by qualitative feedback on usability and user expectations. Findings indicate students face significant SRL challenges, particularly in time management, sustained focus, and independent problem-solving. Participants perceived AI-supported features—such as personalized recommendations, automated feedback, and self-monitoring tools—as beneficial. The system prototype received high usability ratings across platforms, although participants emphasized that AI should complement, not replace, teacher guidance. The Define phase provided critical insights for aligning AI functionalities with SRL principles. This research offers a foundational framework for future development and highlights the importance of learner-centered, ethically grounded AI integration in mathematics education.]]>
