Understanding the Influence of AI-Powered Personalized Learning Systems on Student Engagement and Course Enrollment in Online Education Platforms
TL;DR
The review indicates that while AI personalization is consistently associated with higher engagement and more favorable enrollment intentions in the studies examined, effects are moderated by digital literacy, platform design quality, and institutional context, and the evidence base remains fragmented across disciplines and geographic regions.
Abstract
Abstract Artificial intelligence (AI) has moved from a peripheral support tool to a structural feature of online education, reshaping how content is sequenced, how learners are guided, and how platforms attract and retain enrollment. This paper presents a structured literature review and conceptual analysis of how AI-powered personalized learning systems influence two closely related but distinct outcomes in online education: student engagement and course enrollment decisions. Drawing on peer-reviewed research spanning adaptive learning systems, intelligent tutoring systems, recommender systems, and learning analytics, the paper synthesizes evidence on the mechanisms through which personalization affects behavioral, cognitive, and emotional engagement, and traces the downstream relationship between engagement, perceived value, and enrollment or re-enrollment behavior. A conceptual framework is proposed that links AI personalization features to engagement dimensions and enrollment-related outcomes through mediating constructs such as perceived usefulness, self-efficacy, and course fit. The paper also outlines a proposed mixed-methods research design that future empirical studies could use to test the framework, discusses ethical and equity considerations including data privacy and the digital divide, and identifies gaps in the current literature. The review indicates that while AI personalization is consistently associated with higher engagement and more favorable enrollment intentions in the studies examined, effects are moderated by digital literacy, platform design quality, and institutional context, and the evidence base remains fragmented across disciplines and geographic regions. The paper concludes with implications for online education providers, platform designers, and policymakers, and recommends directions for future longitudinal and cross-institutional research. Keywords: Artificial intelligence; personalized learning; adaptive learning systems; student engagement; course enrollment; online education; recommender systems; learning analytics