Ensuring inclusive digital learning in higher education requires a shift from general principles of accessibility towards a systematic approach that addresses students’ individual educational needs and the barriers they encounter when interacting with digital learning environments. The aim of this study is to develop an empirically grounded conceptual model, INCLUDE-ADAPT, designed to systematically organize identified educational needs and determine potential approaches to adapting digital learning. The methodological framework is based on a problem-oriented analysis of survey responses from higher education students, complemented by descriptive statistical summaries of their responses concerning the organization of learning, digital accessibility, and educational support. The conceptualization of the findings involves systematically linking each identified barrier to the corresponding educational need, a potential adaptive solution, and a mechanism for pedagogical feedback. This approach makes it possible to view inclusivity not merely as the formal availability of digital tools, but as an ongoing process of aligning the characteristics of the educational environment with the diverse needs of its participants. The scientific contribution of this study lies in the conceptual integration of student needs assessment with a systematic approach to identifying and justifying adaptive educational solutions within a unified model. Artificial intelligence is considered a potential tool for supporting personalized learning rather than an empirically validated means of ensuring inclusivity. The proposed model provides a foundation for the further operationalization of its components, the development of evaluation criteria, and the empirical assessment of its applicability across diverse higher education contexts.