Uncovering structural relationships between AI literacy, AI ethics, AI problem-solving, and self-regulated learning

The study aimed to explore how dimensions of AI literacy, specifically, AI technical understanding, AI critical appraisal, and AI practical application affect university students’ AI ethics, their ability to solve problems using AI tools, and their self-regulated learning. Structural equation modelling (SEM) was conducted using AMOS to examine the research model. The findings indicated that AI technical understanding and AI ethics are significant predictors of students’ self-regulated learning. In addition, AI critical appraisal and AI practical application significantly predicted AI ethics, while AI practical application emerged as a significant predictor of AI-supported problem-solving. Other AI literacy dimensions had no significant effects on AI problem-solving and self-regulated learning. The study highlights the necessity of embedding AI literacy into educational policy and curricula not only as a technical skill but also as a multidimensional competence tied to autonomy and ethical reasoning. We emphasize that using AI effectively requires going beyond simple tool proficiency and instead fostering reflective, value-based, and strategic engagement with AI in learning environments. These findings may also contribute to a more nuanced theoretical understanding of AI literacy by identifying components which are most relevant to students’ AI ethics, problem-solving, and self-regulated learning.