<![CDATA[Artificial Intelligence (AI) is increasingly integrated into mathematics education, offering personalized support, adaptive feedback, and interactive learning experiences. However, the mechanisms through which AI-assisted learning improves students’ mathematical problem-solving skills remain underexplored. This study examines the relationships among AI-Assisted Learning (AIL), Self-Regulated Learning (SRL), and Mathematical Problem-Solving Skills (MPSS), with SRL serving as a mediating variable. Drawing on Social Cognitive Theory and Self-Regulated Learning Theory, a quantitative explanatory design was employed. Data were collected from 265 Indonesian secondary school students who had experience using AI-based learning technologies for mathematics learning. The proposed model was analyzed using Partial Least Squares Structural Equation Modeling (PLS-SEM). The findings indicate that AI-Assisted Learning positively and significantly influences Self-Regulated Learning and Mathematical Problem-Solving Skills. Self-Regulated Learning also has a significant positive effect on Mathematical Problem-Solving Skills and mediates the relationship between AI-Assisted Learning and problem-solving performance. These results suggest that the effectiveness of AI in mathematics education depends not only on technological capabilities but also on students’ ability to regulate their learning processes. The study contributes to the Artificial Intelligence in Education literature by identifying self-regulation as a key mechanism linking AI use and higher-order cognitive outcomes. The findings also provide practical insights for educators and developers seeking to design AI-supported learning environments that promote meaningful and autonomous learning.]]>
