<![CDATA[This study aims to examine the effects of generative AI dependency and digital multitasking habits on the quality of self-regulated learning (SRL) among university students, both individually and jointly. The study employed a quantitative approach involving 300 active students from five study programs at the Purwakarta Campus of Universitas Pendidikan Indonesia, who were selected through purposive sampling. Data were collected using four-point Likert-scale instruments that met established validity and reliability criteria. The data were analyzed using multiple linear regression, preceded by classical assumption tests for normality, linearity, multicollinearity, and heteroscedasticity. The results indicated that generative AI dependency was positively and significantly associated with SRL (B = 0.213, p = .004), as was digital multitasking (B = 0.178, p = .009). Simultaneously, both predictors were significantly associated with students’ SRL, F = 27.843, p < .001, and the regression model explained 15.8% of the variance in SRL. These findings suggest that, within the context of this study, moderate levels of generative AI use and digital multitasking did not substantially undermine students’ self-regulatory capacity. Rather than restricting generative AI use, higher education institutions should provide structured guidance on effective AI engagement, including prompt formulation strategies and reflective evaluation of AI-generated outputs, and integrate such guidance into learning task design. Such an approach may help ensure that students’ interaction with generative AI strengthens rather than substitutes for their self-regulatory capacity.]]>
