Drivers of privacy concerns when interacting with a chatbot in a customer service encounter

Abstract

PurposeThis study aims to examine the antecedents of privacy concerns in the era of artificial intelligence. Specifically, it focuses on the impact of various factors related to interactions with a chatbot (creepiness and perceived risk) and individual traits (familiarity with chatbots and need for privacy) in relation to privacy when interacting with a chatbot in the context of financial services. The moderating effect of gender on these relationships was also examined.Design/methodology/approachA total of 430 Canadians responded to an online questionnaire after interacting with a chatbot in the context of a simulated auto insurance quote. A structural equation model was used to test the hypotheses.FindingsThe results showed that privacy concerns are influenced primarily by creepiness, followed by perceived risk and the need for privacy. The last two relationships are moderated by gender. Conversely, familiarity with chatbots does not affect privacy concerns in this context.Originality/valueThis study is the first to consider the influence of creepiness as an antecedent of privacy concerns arising from interactions with AI tools and highlight its key impacts. It also shows how gender moderates specific relationships in this context.