Understanding trust in artificial intelligence: a research agenda
Steven Lockey, Nicole A. Gillespie · Edward Elgar Publishing eBooks · 2024
While the societal and organizational benefits from using Artificial Intelligence (AI) are undeniable, so too are the risks and unique trust challenges, leading to questions about the trustworthiness of these systems and whether they can be trusted. In response to calls to further understand how AI can be developed and deployed in a manner that promotes trust, the literature on trust in AI has burgeoned. In this chapter, we provide an overview of the key limitations and gaps of the literature examining trust in AI and trustworthy AI and outline a future research agenda. We discuss six future research directions to advance a rigorous understanding of trust in AI socio-technical systems: (1) taking a multistakeholder and multilevel approach, (2) understanding optimal trust in AI, (3) examining AI trust failures and repair, (4) closing the gap between trustworthy AI principles and practice, (5) examining the AI trust intention-behavior gap, and (5) taking a contextualized approach to measure and study trust in AI.