The Impact of Error Message Types in AI Financial Chatbots on Usability, Trust, Preference, and Continued Use Intention

As the adoption of AI chatbots continues to expand across the financial industry, the type of message delivered during error situations has been found to influence user experience. This study compared and analyzed the effects of rational and emotional messages in three typical error scenarios of financial chatbots: product guidance, asset inquiry, and loan review. A survey was conducted with 362 individuals who had experience using chatbots, and in-depth interviews were carried out with 10 participants. The analysis revealed that users responded more positively to rational messages that clearly presented the cause of the error and how to resolve it. In particular, statistically significant differences were found in preference and continued use intention depending on the error context. These findings suggest that when designing error messages for financial chatbots, it is important to apply message appeal types appropriately based on the specific context of the situation. This study is meaningful in that it proposes a practical message design strategy to enhance user experience with financial chatbots.