Being Pragmatic About Reliance and Trust in Artificial Intelligence

Abstract The ongoing debate about reliance and trust in artificial intelligence (AI) systems seems to be never ending, challenging our understanding and application of these concepts in human-AI interactions. In this work, we argue for a pragmatic approach to solve this conundrum by focusing on reliance and the three key expectations that should guide human-AI interactions: appropriate reliance, efficiency, and motivation by objective reasons. By focusing on these expectations, we show that it is possible to reconcile reliance with trust in a manner that is both theoretically sound and practically useful. As it turns out, reliance is the key relation of interest while trust in AI is a derived concept that helps explaining these expectations. Our reliance-centered framework does not dismiss the concept of trust in AI but repositions it as a key property of reliance, offering a pragmatic alternative to classical rational or motivational accounts of trust that prove difficult to apply in the context of human-AI interactions. As AI continues to integrate into society, particularly in high-stakes environments like healthcare, our pragmatic approach provides a practical and meaningful framework for addressing the nuances of trust in AI.