The Trust Threshold: How Public Perception of AI Harm Moderates the Impact of FinTech Innovation on Systemic Banking Stability

Abstract

The rapid integration of artificial intelligence into financial services has ushered in an era of unprecedented innovation, yet it has simultaneously introduced complex socio-technical risks that challenge conventional understandings of banking stability. While substantial research has examined the technical dimensions of AI model governance and regulatory compliance, comparatively little attention has been devoted to understanding how public perceptions of AI-related harm influence the relationship between technological innovation and systemic financial resilience. This study addresses this critical gap by investigating the moderating role of public trust and perceived socio-political harm in the innovation-stability nexus within the banking sector. Drawing upon diffusion of innovation theory and systemic risk frameworks, we develop and test a conceptual model wherein public perception of AI harm—encompassing concerns regarding algorithmic fairness, data privacy, and opaque decision-making—moderates the impact of aggressive AI adoption on long-term banking stability. The research employs a mixed-methods approach, combining longitudinal analysis of banking stability indicators across major financial institutions with survey data capturing public sentiment toward AI deployment in financial services. Preliminary findings suggest that while AI innovation initially enhances operational efficiency and profitability, these benefits are contingent upon maintaining public confidence in the fairness and integrity of automated systems. When perceptions of harm exceed a critical threshold, the stability benefits of innovation diminish significantly, potentially triggering customer attrition, regulatory intervention, and systemic contagion effects. The study draws upon foundational insights from Cao et al. (2021), who examined consumer trust dynamics in algorithmic financial advising, and extends the work of König et al. (2022), who explored the reputational contagion mechanisms linking perceived AI failures to broader institutional stability. By illuminating the psychological and sociological dimensions of AI governance, this research contributes to emerging scholarships on trustworthy AI and offers practical guidance for financial institutions seeking to balance innovation imperatives with the maintenance of public trust. The findings underscore the necessity of embedding public perception monitoring into systemic risk assessment frameworks and highlight the importance of transparent, explainable AI architectures in preserving the social license upon which banking stability ultimately depends.