Personalized but Valued? How AI-Driven Advertising Shapes Perceived Trust and Purchase Intention: A Cross-National Comparison

AI-driven personalization has become a defining feature of digital advertising, yet whether it drives purchase intention directly or depends on other underlying mechanisms remains insufficiently understood, particularly in under-studied emerging markets. This study investigates how perceived relevance, usefulness, and privacy concerns shape perceived trust and purchase intention in response to AI-driven personalized advertising across Lebanon and Türkiye, two digitally distinct markets. Drawing on the Stimulus Organism Response model, Technology Acceptance Model, Privacy Calculus Theory, and Trust in Automation Theory, a quantitative cross-national design was employed, using a structured questionnaire administered to 802 social media users (394 Lebanon; 408 Türkiye), with data analyzed through partial least squares structural equation modeling and permutation-based multigroup analysis following partial measurement invariance. Results confirm perceived personalization has no direct effect on purchase intention, operating solely through relevance, usefulness, and privacy concerns; relevance and usefulness build trust while privacy concerns erode it, and trust, relevance, and usefulness each independently drive purchase intention. This study advances an asymmetric mediation account of AI advertising, positioning trust as the conduit through which privacy risk reaches behavior and, for managers, the lever converting personalization into purchase. Future research could examine these dynamics longitudinally or across other digitally emerging markets.