Breaking Platform Silos in Digital Higher Education: A Standards-Based Interoperability Framework for Learning Management Systems, Student Information Systems, and AI Services

Universities now run teaching and administration on many separate platforms. A learning management system (LMS) holds course content and assignment scores, a student information system (SIS) holds enrolments and official grades, and a growing number of AI services, from tutoring chatbots to feedback tools, each need part of that data to work. When these systems are connected through manual exports or one-off custom code, staff enter the same data twice, records disagree, data protection is weak, and AI tools either see too little to be useful or far more than they should. This paper proposes a standards-based interoperability framework that connects LMS, SIS and AI services in higher education. Following design science research, the study analyses current open specifications, including LTI 1.3, OneRoster 1.2, Edu-API, Caliper Analytics 1.2, xAPI 2.0, Open Badges 3.0, the Comprehensive Learner Record 2.0 and W3C Verifiable Credentials 2.0, together with two new protocols for AI systems, the Model Context Protocol and Agent2Agent. The resulting framework has four parts: a layered reference architecture, a system-of-record matrix that names one authoritative source for each data object, an AI gateway that gives AI services only course-scoped context, and a five-level maturity model. The framework is mapped to the four layers of the European Interoperability Framework and illustrated with an end-of-semester grade and credential flow set in the context of Uzbekistan's national higher education information system. The paper also discusses governance, data localisation and security, and outlines steps for empirical validation.