Treating differently or equally: A study exploring attitudes towards AI moral advisors

Abstract

Artificial intelligence (AI) technology has evolved from serving primarily as a decision-maker in the past to increasingly taking on the role of an advisor. However, contemporary attitudes toward AI applications in moral decision-making remain unclear. In Study 1, we explored whether there is a difference in attitudes towards human and AI moral advisors when both are defined in the context of a one-time decision-making scenario. Studies 2a and 2b, with the goal of achieving higher ecological validity, optimized decision-making methods and scenarios, respectively. We obtained consistent results, indicating that people equally trust the advice of AI moral advisors and human moral advisors. When it comes to assigning responsibility after a decision, individuals assign responsibility equally to both AI and human advisors.