AI Utilization Strategies for Personalized Learning in K-12

Recent advances in artificial intelligence (AI) have expanded the possibilities of personalized learning in K-12 education by enabling real-time learner data analysis and adaptive instructional support. However, previous studies have mainly examined the effectiveness of individual AI tools, with limited attention to how AI can be systematically used from an instructional design perspective. This study, therefore, aimed to derive and conceptualize AI utilization strategies for personalized learning in K-12 education. A systematic literature review was conducted following the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) 2020 guidelines. Empirical studies on AI-based personalized learning in K-12 education published between 2010 and 2025 were collected from major academic databases. Tomlinson’s Content–Process–Product framework was used to extract and categorize AI utilization strategies, and the categorization criteria were refined through iterative consensus-building procedures. The analysis yielded seventeen AI utilization strategies across the three dimensions of Content, Process, and Product. In the Content dimension, AI adjusts the difficulty, structure, representation, and context of learning materials. In the Process dimension, it supports learning pathways, scaffolding, interaction, and self-regulation. In the Product dimension, it enables continuous diagnosis, feedback, and visualization of learning growth. These strategies were also, at times, applied in interlocking combinations within a single design. These findings suggest that AI-based personalized learning can be used not only as individual strategies but also designed as an integrated instructional system. This study provides an analytical framework for future research, educational practice, and policy development.