Exploring Personalized Explanation Methods in Fintech Decision-Making Chatbot Based on User Characteristics Using Large Language Model

As AI increasingly supports financial decision-making, the need for explanations that users can understand and trust is growing. Previous research has often taken a uniform approach to explanations, without considering individual user characteristics. In this work, we examine the interaction between explanation methods and user characteristics and propose personalized explanations. To identify suitable explanations for different levels of risk-taking, Study 1 employed a 3 (explanation methods) × 2 (risk-taking levels) between-subjects design. We also found that subjective AI knowledge plays an important role and proposed corresponding explanations. Study 2 validated the explanation methods identified in Study 1 using a 2 (risk-taking level–based explanations) × 2 (AI literacy–based explanations) between-subjects design. The results showed that personalized explanations improved user experience. Furthermore, we examined whether financial literacy moderates the relationship between explanation methods and user experience, finding that it moderates the effect of AI literacy–based explanations on understanding.