<![CDATA[AI-based recommender systems are widely implemented in e-commerce and digital platforms due to their ability to provide relevant and personalized recommendations. However, a key challenge in sustaining their adoption is establishing user trust, as limited trust can reduce continued usage intentions. This study investigates the effects of interactivity and personalization on intention to use, with trust as a mediating variable. A quantitative correlational approach was applied using Partial Least Squares Structural Equation Modeling (PLS-SEM). Data were collected from 170 Universitas Kuningan students who actively used recommender-based applications such as Shopee, Tokopedia, Netflix, and Spotify. The findings indicate that interactivity (β = 0.059; p = 0.493) and personalization (β = 0.210; p = 0.053) do not directly influence intention to use. Nevertheless, both interactivity (β = 0.233; p = 0.003) and personalization (β = 0.571; p < 0.001) significantly affect trust. Furthermore, trust has a strong positive impact on intention to use (β = 0.525; p < 0.001) and significantly mediates the relationships between interactivity and personalization with intention to use. The model explains 55.4% of the variance in trust and 53.7% in intention to use. These results emphasize trust as a critical mechanism linking user experience to continued use of AI-based recommender systems.]]>
Copyrights © 2025
