Abstract:

This work-in-progress paper examines the development of artificial intelligence (AI) tools to support self-paced learning and address persistent challenges in education. Traditional teaching methods with fixed classroom schedules and uniform pacing, struggle to meet the diverse needs of students leading to gaps in understanding and engagement, such as, many learners fall behind or disengage altogether, unable to find alignment between the curriculum and their individual learning trajectories. Emerging AI technologies, personalize instruction based on individual student performance, show potential to offer a path forward by adjusting the pace, sequence, and delivery of content in real time, which are promising to close the learning gaps and sustain motivation. While early models, such as intelligent tutoring systems, have demonstrated encouraging results, critical questions remain about their effectiveness, practical implementation and broader implications in education. This study surveys recent research on AI-enabled self-paced learning, with a focus on machine learning-based personalization and its effect on student engagement and outcomes. Key questions explored include: how effectively do these systems adapt to different learner profiles? To what extent do they influence student persistence, comprehension, and interest? And what structural or cultural barriers stand in the way of broader adoption? The review categorizes findings by underlying AI methodologies and the nature of reported educational gains. While initial evidence supports the benefits of personalization, particularly in online and asynchronous learning environments, important limitations persist. Issues around algorithmic transparency, data privacy, and integration into traditional classroom contexts remain unresolved. Notably, most of the existing literature focuses on K–12 settings or massive open online courses (MOOCs), with far less attention given to traditional higher education. This gap suggests an urgent need for research into how AI-driven self-paced learning might function in structured, instructor-led university environments. Future work identified in this paper includes the development of robust evaluation frameworks, principles for ethical and inclusive implementation, and strategies that combine AI technologies with teacher expertise rather than attempting to replace it. Ultimately, the goal is to automate education and support more flexible, responsive, and humane learning environments that help students move forward with confidence.

Keywords:

Artificial intelligence in education, adaptive learning, personalized learning, self-paced learning, intelligent tutoring systems, machine learning in education.