The rapid integration of Artificial Intelligence (AI) into the higher education ecosystem has fundamentally disrupted traditional paradigms of knowledge production and dissemination. While AI adoption rates among students have surpassed 90%, the absence of a coherent and native governance model in Iran has exacerbated risks related to the digital divide and ethical challenges. The primary objective of this research is to design a strategic framework for AI governance within Iran's higher education system. The research methodology is a "qualitative comparative study" utilizing a "three-level analytical framework" (Macro, Meso, and Micro) to examine the experiences of three selected countries: Singapore, India, and Germany. Findings indicate that global governance approaches can be categorized into three distinct typologies: 1) The "Centralized-Instrumental" model (Singapore), emphasizing efficiency and trust by design; 2) The "Developmental-Market" model (India), focusing on scalability and inclusion; and 3) The "Legal-Federal" model (Germany), centered on ethics and human dignity. Finally, by synthesizing these experiences and aligning them with Iran's upstream documents, a hybrid policy package is proposed. This model rests on three pillars: "Intelligent Authority in Regulation," "Digital Justice in Access," and "Ethical Protection of Scientific Authority." Key implementation strategies include establishing a National AI Commission (Macro level), creating AI Innovation Offices in universities (Meso level), and launching an AI Literacy Movement (Micro level).