Artificial intelligence is playing a growing role in consumer financial services, yet many users lack the expertise to judge AI-generated advice and therefore rely on the explanations provided. The present study investigates how explanation quality, advisor type, and decision risk jointly shape users’ evaluation and adoption of financial recommendations. Building on the Information Adoption Model and explainable AI literature, we propose a framework where information quality, information usefulness, perceived trust, and perceived risk influence intentions to follow advice. Conducting a controlled online experiment, we also examine how these relationships may differ across advisor types (human, low-explainability AI, high-explainability AI) and decision risk levels (low, high). Results show that the effects of information usefulness, trust, and information quality differ systematically depending on advisor type, explanation clarity, and decision risk level. Our findings demonstrate that information adoption arises from the interaction of informational, relational, and contextual factors, extending the Information Adoption Model and offering insights for designing more effective AI–human financial advisory systems.
Ioannou et al. (Fri,) studied this question.