The rapid integration of AI-powered recommendation engines into e-commerce platforms has transformed how consumers interact with digital content and make purchase decisions. This study explores the behavioral and psychological mechanisms underpinning consumer responses to AI-generated product suggestions, emphasizing the roles of perceived personalization, algorithmic trust, and perceived intrusiveness. Anchored in dual-process theory, the research employed a cross-sectional design with 430 online shoppers, analyzed through PLS-SEM and multiple regression techniques.

Findings indicate that personalization (β = 0.381, p < 0.001) and trust (β = 0.276, p < 0.001) exert significant positive effects on purchase intention. Conversely, perceived intrusiveness has a negative impact (β = –0.218, p < 0.001). Mediation analysis further revealed that intrusiveness partially mediates the effects of both personalization and trust on purchase decisions. Subgroup analyses showed generational differences, where Gen Z consumers place greater emphasis on personalization, and platform-inexperienced users depend more heavily on algorithmic trust. The model demonstrates strong explanatory power (R² = 0.633) and predictive relevance (Q² = 0.402), underscoring its robustness.

These results underscore the necessity for brands to design AI systems that are not only efficient but also psychologically attuned to user expectations. Transparent and ethically sound personalization strategies that mitigate perceived intrusiveness can significantly enhance consumer trust and decision-making outcomes. The study offers practical implications for marketers, AI developers, and digital policy advocates in optimizing consumer engagement in data-driven environments.