Moral Judgment in the Age of Artificial Intelligence: How Humans Perceive Ethical Decisions Made by Machines

Artificial intelligence (AI) poses new challenges to human rationality and moral judgment in a volatile, uncertain, complex, and ambiguous (VUCA) world. This paper integrates key insights from moral philosophy, normative and applied ethics, and management studies and proposes a behavioral perspective on the ethics of AI. Prior research has indicated that people have biased perceptions towards AI’s decision outcomes; however, less attention has been paid to understanding human rationality and moral judgment under the complexity and ambiguity of the AI age. This paper focuses on people’s perception of the ethicality of decisions made by human versus AI systems. Specifically, it examines the perception of identical (un)ethical decisions made by human versus AI systems and decisions applying different moral principles (Utilitarian vs. Kantian) in a moral dilemma. I argue that humans are perceived as more unethical than AI when both make the same unethical decision, yet paradoxically, human decisions are perceived as more ethical than identical ones made by AI. Moreover, when moral dilemmas involve different ethical principles, utilitarian decisions made by humans are judged as less ethical than those made by AI. Conversely, non-utilitarian indecision lacking clear reasoning is perceived as less ethical for AI than for humans, although this difference disappears when the non-utilitarian decision is grounded in Kantian reasoning. By proposing a behavioral perspective on the ethics of AI, this paper contributes to a more comprehensive understanding of bounded human rationality and moral perception in the age of artificial intelligence.