Outlines a conceptual framework unifying research on attitudes toward AI based on the premise that the decision to use a predictive algorithm is primarily shaped by users’ expectations of its performance.