Navigating Risk and Trust in AI Financial Systems: The Role of Institutional Structures and the EU AI Act

Abstract

This study investigates how institutional structures influence trust in artificial intelligence (AI) financial systems, focusing on the implications of the EU AI Act. Drawing on institutional theory and institution-based trust, a survey of 220 financial services consumers in EU countries examined how institutional signals affect perceived risk and trust. Structural equation modeling revealed that robust institutional structures directly increase trust and indirectly enhance it by reducing perceived risk. Perceived risk, in turn, negatively influences trust in AI systems. By clarifying these pathways, the study extends prior work to test a theory-driven model of trust formation. The findings highlight that institutional structures act not only as safeguards but also as legitimacy signals, shaping user confidence in high-stakes AI environments. These insights are particularly relevant for policymakers and financial institutions aiming to foster public trust in AI technologies under evolving regulatory regimes like the EU AI Act.