Understanding the determinants of the outcomes for artificial intelligence in personalized learning: the rule of task completion time

Abstract

Artificial intelligence (AI) for personalized learning is an emerging field that has attracted growing attention from educational technology researchers. Recent developments have emphasized the importance of system use, user satisfaction, and educational outcomes. We propose a model combining the DeLone & McLean IS success model and the belief-actionoutcome (B-A-O) model to evaluate the impact of AI-based personalized learning systems on both humanistic and instrumental outcomes. The empirical approach is based on data collected from 301 student interviews across diverse educational settings. The results reveal that use and user satisfaction are critical antecedents of perceived task performance, individual performance, and quality of life. Furthermore, we demonstrate the moderating role of task completion time, which significantly influences the strength of the relationships between system use, satisfaction, and the final outcomes. Understanding the influence of AIdriven learning platforms on student performance and well-being offers valuable insights for developers, educators, and policymakers aiming to optimize adaptive educational technologies. We provide both theoretical and practical implications based on our findings.