The rapid adoption of generative artificial intelligence (GenAI) tools such as ChatGPT among university students has outpaced the field's understanding of their psychological consequences. While much attention has focused on academic integrity and technological capability, comparatively less systematic attention has been given to how GenAI use shapes students' self-beliefs, self-regulatory capacities, and psychological wellbeing, questions that sit squarely within the domain of educational psychology. This study employs a qualitative library research (studi pustaka) design to synthesize empirical and theoretical literature around three psychological foci: (1) self-efficacy and academic motivation among GenAI-using students; (2) self-regulated learning and the risk of cognitive offloading arising from GenAI dependency; and (3) psychological wellbeing and academic self-control among GenAI users. Sources were drawn from peer-reviewed empirical studies, foundational social-cognitive theory, and institutional guidance published between 1977 and 2025, analyzed thematically. Findings indicate that GenAI use is empirically associated with elevated academic self-efficacy, consistent with Bandura's social-cognitive theory, yet this same ease of use fosters cognitive offloading that experimental evidence links to "metacognitive laziness" and weaker self-regulated learning processes. Furthermore, a large-sample mediation study found that higher GenAI usage was significantly associated with lower self-control and lower academic wellbeing, with self-control partially mediating this relationship. The study concludes that GenAI functions as a genuine psychological double-edged sword for students: it can strengthen momentary confidence and motivation while simultaneously eroding the self-regulatory and self-controlling capacities on which durable academic wellbeing depends, making the deliberate cultivation of metacognitive and self-regulatory skills an urgent priority for educational psychology practice in higher education.
Copyrights © 2026