Most people would trust and accept judges using AI if they became victim of a crime.

What is it about?

We asked participants to imagine being the victim of a street crime and presented them with court proceedings in which we manipulated the judge’s use of generative AI (no use vs. use). We investigated whether participants evaluated the judge and the verdict differently depending on the judge’s use of generative AI. We found that the use of generative AI only partially reduced the perceived trustworthiness of the judge, as he was perceived to have less integrity. However, the use of AI had no effect on perceived competence, benevolence, trust in the judge, acceptance of the verdict, or perceived interpersonal justice.

Featured Image

Photo by Wesley Tingey on Unsplash

Photo by Wesley Tingey on Unsplash

Why is it important?

The use of generative AI in judicial decision-making is on the rise worldwide. However, it is not known how potential victims perceive the use of generative AI in morally sensitive contexts with high stakes. For example, people might think that judges who use generative AI to reach a verdict are not up to the task, are lazy, and do not care about the victim. Studies on AI use in medical diagnostics indicate these effects for the perception of physicians. Our findings show that the use of generative AI by judges affects only perceived integrity; we demonstrate that the negative effects are limited, thereby highlighting the potential for using generative AI in judicial decision-making while maintaining public trust and acceptance.

Perspectives