Will Customers’ Understanding of the Trolley Dilemma Hinder Their Adoption of Robotaxi?

Abstract

Robotaxi, coined from “robot” and “taxi”, refers to a taxi service with vehicles controlled by self-driving algorithms instead of human drivers. Despite the availability of such a service, it is yet unknown whether customers will adopt robotaxi, given its immaturity. Meanwhile, the potential customers of the robotaxi service are facing an inescapable ethics issue, the “trolley dilemma”, which might have a strong impact on their adoption of the service. Based on the necessity of understanding robotaxi adoption, especially from an ethical point of view, this study aims to uncover and quantify the antecedents of robotaxi adoption, taking the trolley dilemma into consideration. We applied a modified Unified Theory of Acceptance and Use of Technology (UTAUT) framework to explore the antecedents of robotaxi adoption, with a special focus on customers’ understanding of the trolley dilemma. We conducted online surveys (N = 299) to obtain the customers’ opinions regarding robotaxis. Aside from measuring standard variables in UTAUT, we developed four proprietary items to measure trolley dilemma relevance. We also randomly assigned the participants to two groups, either group A or group B. Participants in group A are told that all robotaxis are programmed with a utilitarian algorithm, such that when facing a trolley dilemma, the robotaxi will conditionally compromise the passenger(s) to save a significantly larger group of pedestrians. In the meantime, participants in group B are informed that all robotaxis are programmed with an egocentric algorithm, such that when facing a trolley dilemma, the robotaxi will always prioritize the safety of the passenger(s). Our findings suggest that both performance expectancy and effort expectancy have a positive influence on robotaxi adoption intention. As for the trolley dilemma, customers regard it as of high relevance to robotaxis. Moreover, if the robotaxi is programmed with an egocentric algorithm, the customers are significantly more willing to adopt the service. Our paper contributes to both adoption studies and ethics studies. We add to UTAUT two new constructs, namely trolley dilemma relevance and trolley dilemma algorithm, which can be generalized to adapt to other new technologies involving ethics issues. We also directly ask customers to assess the relevance and algorithm of the trolley dilemma, which is a meaningful supplement to existing ethics studies that mostly debate from researchers’ perspectives. Meanwhile, our paper is managerially meaningful as it provides solid suggestions for robotaxi companies’ marketing campaigns.