Sharing information with AI (versus a human) impairs brand trust: The role of audience size inferences and sense of exploitation

Abstract

This research examines whether and why disclosing information to AI as opposed to humans influences an important brand-related outcome—consumers’ trust in brands. Results from two pilot studies and nine controlled experiments (n = 2,887) show that consumers trust brands less when they disclose information to AI as opposed to humans. The effect is driven by consumers’ inference that AI shares information with a larger audience, which increases consumers’ sense of exploitation. This, in turn, decreases their trust in brands. In line with our theorizing, the effect is stronger among consumers who are relatively more concerned about the privacy of their data. Furthermore, the negative consequences for brands can be mitigated when (1) customers are informed that the confidentiality of their information is protected, (2) AI is anthropomorphized, and (3) the disclosed information is relatively less relevant.