Blaming Humans and Machines: What Shapes People’s Reactions to Algorithmic Harm

Artificial intelligence (AI) systems can cause harm to people. This research\nexamines how individuals react to such harm through the lens of blame. Building\nupon research suggesting that people blame AI systems, we investigated how\nseveral factors influence people's reactive attitudes towards machines,\ndesigners, and users. The results of three studies (N = 1,153) indicate\ndifferences in how blame is attributed to these actors. Whether AI systems were\nexplainable did not impact blame directed at them, their developers, and their\nusers. Considerations about fairness and harmfulness increased blame towards\ndesigners and users but had little to no effect on judgments of AI systems.\nInstead, what determined people's reactive attitudes towards machines was\nwhether people thought blaming them would be a suitable response to algorithmic\nharm. We discuss implications, such as how future decisions about including AI\nsystems in the social and moral spheres will shape laypeople's reactions to\nAI-caused harm.\n