Artificial intelligence (AI) has been a transformational force in higher education today, providing new opportunities to differentiate instruction to meet individual learning needs. Despite the creation of multiple online platforms, digital assessments and automated feedback systems, many universities still use traditional, uniform teaching practices that do not cater for different learner backgrounds, motivations and cognitive preferences. Although AI-powered personalization is widely acknowledged as an exciting solution, empirical evidence on how well it works, especially in real classroom settings and in developing regions is scarce. This study addresses this gap by assessing the impact of an AI-based adaptive learning system that dynamically aligns the instructional content according to the performance trend, pace of learning and engagement behaviors of the students. The research will be conducted using a mixed-methods methodology, which will combine performance metrics and analytics, tailored content recommendations, engagement metrics, and surveys of student perceptions to analyze both the quantitative increase in learning outcomes and qualitative information about the perceptions of the learner. Findings show that AI-powered personalization has an important impact on improving academic performance, engagement, and be well received by students as supportive and accessible. The study adds to the practical and theoretical implications for the institutions who are interested in incorporating adaptive learning technologies, providing evidence-based recommendations for effective adoption in higher education settings.