Linking Privacy Concerns with Resistance to Chatbotsin the Banking Sector: Creepiness as a Mediator andNeed for Human Interaction as a Moderator

Abstract

Despite the Popularity of Artificial Intelligence, consumers resist using AI-based virtual agents. Only a few limited research studies have identified the reasons for users’ negative  experience with chatbots. The study’s objective was to uncover the underlying factor of customer resistance towards the chatbots of the banking sector. The study employed the SOR framework due to its low application in determining the negative user experience in chatbot studies. The research model hypothesized the relationship between privacy concerns (stimulus), creepiness (organism), and resistance toward chatbots (response). The need for human interaction was taken as a moderator. Data were collected from chatbot users by using an online survey questionnaire. Data were analyzed by using covariance-based structural equation modeling. The results confirmed the positive relationship between privacy concerns and customers’ resistance towards chatbots. The mediation effect of creepiness was also supported; however, the need for human interaction as a moderator was not accepted as hypothesized. The study findings offer valuable insights for marketers, practitioners, and policymakers.

Keywords: privacy concerns, creepiness, resistance and chatbots