One of the great challenges in education is student learning, where everyone has a different learning style. The use of Artificial Intelligence (AI) algorithms to recommend learning materials has been explored in the academic world. This article investigates the main AI algorithms used for learning recommendation, addressing their potential contributions to teaching personalization, adaptive learning, student support, research, data analysis and automatic feedback generation. The objectives of this paper are to present these potentialities by systematically collecting and analyzing the studies available in literature to: (I) gain a comprehensive understanding of the interconnections between AI and education; (II) provide an overview of the current state of research on this topic; and (III) identify important gaps in existing approaches as well as promising research trends. To achieve these objectives, a systematic literature review was carried out covering articles published in journals from three relevant databases. Initially, 99 primary studies were selected, which underwent a selection and peer review process, resulting in 62 final articles, whose models and solutions for using AI in learning recommendation were classified and summarized. The results of the research show the main AI algorithms used for learning recommendation in education, and we can highlight Support Vector Machine, Naïve Bayes, Decision Tree which had the highest concentration of articles.