THE AI TRUST DILEMMA: UNRAVELING THE PSYCHOLOGICAL BARRIERS TO ARTIFICIAL INTELLIGENCE ADOPTION

Abstract

The purpose of this study is to empirically examine the impact of perceived artificial intelligence (AI) related risks, specifically privacy risk, algorithmic bias, AI hallucination, and job displacement fear, on consumer trust and subsequent behavioral intentions toward AI technologies. Drawing upon the Technology Acceptance Model (TAM) and Risk-Trust frameworks, this study adopts a quantitative, cross-sectional design. Primary survey data were collected from 250 digitally active consumers in Pakistan with prior experience utilizing AI-enabled services. Direct effects and trust-mediated relationships were evaluated using multiple regression and mediation analysis. The empirical results demonstrate that all four risk dimensions significantly and negatively affect consumer trust in AI. AI hallucination exerts the strongest trust-eroding effect, followed closely by privacy risk, algorithmic bias, and job displacement fear. Furthermore, consumer trust exerts a strong positive influence on behavioral intention and partially mediates the adverse impacts of perceived AI risks on adoption intentions. This research addresses a significant geographic gap by providing evidence from the Pakistani consumer market, offering critical theoretical extensions and practical guidance for practitioners and policymakers navigating emerging digital economies.