Trust and Fraud in Decentralized Digital Transactions: A Structural Equation Modeling Approach Integrating Blockchain and Artificial Intelligence

Abstract

Decentralized digital transactions, such as those occurring via blockchain-enabled platforms and AI-driven systems, are rapidly transforming financial and commercial ecosystems. However, user participation is often constrained by concerns over fraud and a lack of trust. This study develops and empirically tests a structural equation modeling (SEM) framework that investigates how perceived blockchain transparency and perceived AI competence and transparency shape trust in platforms, perceived fraud risk, and users' transaction intentions. Grounded in trust-risk theory, socio-technical systems theory, and technology acceptance models, the framework incorporates five core constructs. Data collected from 412 active users of decentralized platforms were analyzed using SEM to validate the hypothesized relationships. Findings reveal that perceived blockchain transparency and AI competence significantly enhance trust while simultaneously reducing perceived fraud risk. Moreover, trust in the platform positively influences transaction intention, whereas perceived fraud risk has a negative impact on both trust and intention. The study contributes to the literature on digital trust and decentralized finance by integrating socio-technical factors. Practical implications are offered for platform designers and policymakers seeking to build trustworthy and fraud-resilient systems using blockchain and AI technologies.