In contrast to system control, learner control over the selection of learning tasks enables learners to practice self-directed learning skills. Paradoxically, self-directed learning skills are yet prerequisite for effectively controlling and personalizing one's own learning. This is called the self-directed learning paradox. This article presents an overview of different approaches to personalized learning and explores whether shared control, in which the system preselects a subset of optimal tasks and the learner makes the final task selection, offers a way out to this paradox. Furthermore, this article provides guidelines for instructional design based on the concept of shared control. Finally, limitations of the model are acknowledged and theoretical implications for future studies are discussed.