Relationship between artificial intelligence tool usage experience and academic stress among college students: Mediating role of loneliness and moderating role of academic self-efficacy

As artificial intelligence (AI) rapidly integrates into higher education, AI tools are increasingly being utilized to support student learning. Although these tools offer efficiency and convenience, their psychological implications-particularly vis-à-vis academic stress-remain unclear. This study investigated the relationship between AI tool usage experience and academic stress among college students, focusing on the potential mediating role of loneliness and the moderating role of academic self-efficacy. Overall, 624 university students were surveyed using the AI Tool Usage Experience Scale, UCLA Loneliness Scale, Academic Stress Scale, and Academic Self-Efficacy Scale. The following three key findings were observed: (1) AI tool usage experience significantly positively predicted students' academic stress. (2) Loneliness partially mediated this relationship. (3) Academic self-efficacy significantly moderated the mediation pathway's first stage. Specifically, AI usage's positive predictive effect on loneliness was stronger (weaker) for students with higher (lower) academic self-efficacy levels. These findings suggest that AI tool usage not only directly influences academic stress but also contributes indirectly through heightened feelings of loneliness, particularly among students with strong self-efficacy beliefs. This study underscores the complex psychological mechanisms underlying students' interactions with AI in educational settings.