Data_Sheet_1_Artificial Intelligence Can’t Be Charmed: The Effects of Impartiality on Laypeople’s Algorithmic Preferences.docx

Over the coming years, AI could increasingly replace humans for making complex decisions because of the promise it holds for standardizing and debiasing decision-making procedures. Despite intense debates regarding algorithmic fairness, little research has examined how laypeople react when resource-allocation decisions are turned over to AI. We address this question by examining the role of perceived impartiality as a factor that can influence the acceptance of AI as a replacement for human decision-makers. We posit that laypeople attribute greater impartiality to AI than human decision-makers. Our investigation shows that people value impartiality in decision procedures that concern the allocation of scarce resources and that people perceive AI as more capable of impartiality than humans. Yet, paradoxically, laypeople prefer human decision-makers in allocation decisions. This preference reverses when potential human biases are made salient. The findings highlight the importance of impartiality in AI and thus hold implications for the design of policy measures.

在未来数年，人工智能有望在制定复杂决策方面逐步取代人类，这得益于其在标准化和去偏决策程序方面的承诺。尽管关于算法公平性的争论激烈，但鲜有研究探讨当资源分配决策权移交至人工智能时，公众如何作出反应。本研究通过考察感知的公正性作为影响人工智能作为人类决策者替代品的接受度的因素，来回答这一问题。我们提出，公众倾向于认为人工智能比人类决策者更加公正。我们的研究表明，人们在涉及稀缺资源分配的决策程序中重视公正性，并且人们认为人工智能在公正性方面比人类更具能力。然而，出人意料的是，在资源分配决策中，公众更倾向于选择人类决策者。当潜在的人类偏见被凸显时，这一偏好发生逆转。研究结果突显了人工智能公正性的重要性，从而对政策措施的设计具有深远影响。