Do consumers’ perceptions of algorithms and trusting beliefs in providers affect perceived structural assurances of AI-powered applications?

Abstract

This study aims to understand how perceptions of algorithms and trusting beliefs in service providers facilitate consumers’ perceived structural assurance of using commercial AI applications. The present study adopts a combined approach of partial least squares-structural equation modeling and fuzzy-set qualitative comparative analysis (PLS-SEM-fsQCA) to understand the linear and combined effects of the studied factors on perceived structural assurance with 297 effective responses. The PLS-SEM findings revealed that algorithmic perceptions (i.e., Fairness, Accountability, and Transparency) and trusting beliefs (i.e., Benevolence, Competence, and Integrity) were positively associated with Perceived Structural Assurance. The fsQCA findings indicate four configural solutions of causal conditions that explain Perceived Structural Assurance, and each solution reflects a particular type of consumers who have unique considerations when assessing commercial AI’s structural assurance. This study adds to consumer behavior studies by introducing consumers’ perceptions of algorithms and trusting beliefs in evaluating their structural assurances in commercial AI applications from linear and complexity perspectives.