Artificially Intelligent Customer Service: Marketplace Implications and Consequences

This dissertation explores how emergent technologies influence consumer experience and market behavior. The first chapter of the dissertation, “The Effect of Implementing Chatbot Customer Service on Stock Returns: An Event Study Analysis,” employs the event study method to examine how the implementation of customer service chatbots impacts firm value. I find that investors respond positively to chatbot implementation and that Business-to-Business (B2B) companies have significantly more to gain from this service innovation compared to Business-to-Consumer (B2C) companies. However, anthropomorphizing chatbots attenuates this positive financial impact for B2B companies, suggesting differences in expectations and preferences of B2C and B2B customers. A survey of investors and a consumer experiment provide additional evidence for these relationships. While prior research suggests that Artificial Intelligence (AI) implementation may have a negative impact on firm value, using the example of AI chatbots, this research provides the first evidence that investors respond favorably to a customer-centric application of AI. In addition, it broadens the predominant consumer-focused lens of the marketing literature on AI to include B2B customer and investor perspectives. Further, while marketing research has primarily viewed anthropomorphism as a categorical variable, I propose an anthropomorphism index that offers a more nuanced approach to examining the anthropomorphism of AI technology.In the second chapter, “Understanding Antagonistic Consumer Behavior Toward Humanlike Robots in the Marketplace,” I examine consumer antagonism toward service robots and contribute theoretically and substantively to the emerging marketing literature on AI. Using carefully designed experiments, I demonstrate that consumers tend to dehumanize and, consequently, behave antagonistically toward humanoid (vs. non-humanoid) service robots. While prior research shows that people avoid humanoid robots, this research is the first to show that consumers dehumanize and subsequently engage in anti-normative, negative approach behavior when such robots are deployed in customer-facing service roles. Furthermore, this research contributes to the dehumanization literature by demonstrating that dehumanization is a two-step process when applied to service technology. Importantly, the findings help reconcile the contradiction between the literature on anthropomorphism and the uncanny valley hypothesis by demonstrating that consumers receive humanoid (vs. non-humanoid) robots more positively when the effect of dehumanization is mitigated.