Investor Trust in Robo-Advisors: The Impact of Perceived Risk, Control, and AI Predictive Capability

Abstract

This research investigates how perceived risks to data security and perceived control influence investors' trust in robo-advisors, as well as the impact of this trust on their adoption decisions. It also examined the moderating effect of the predictive advantage of machine learning. Employing a quantitative analysis method with Partial Least Squares Structural Equation Modeling (PLS-SEM), data were gathered from investors who are familiar with robo-advisors. The results indicate that perceived data security risk diminishes trust, whereas perceived control enhances trust. However, trust alone does not directly result in adoption, implying that other elements such as ease of use and economic benefits might be influential. The predictive advantage of machine learning enhances the link between trust and adoption but does not directly affect adoption decisions. Providers of robo-advisors should focus on improving algorithm transparency, bolstering data security systems, and offering features that enhance users' perceived control. Effective communication AI's predictive advantages of AI is also essential for building investor trust and alleviating security concerns.