Why does algorithmic transparency matter for policy acceptance? The mediating role of intent attributions under varying organizational reputations

Artificial intelligence (AI) is increasingly being integrated into public service operations, yet the success of AI-enabled governance, particularly in mandatory surveillance contexts, hinges critically on public attitudes towards AI initiatives. Drawing on signalling theory and attribution theory, this study identifies intent attribution as a novel cognitive mechanism by examining how individuals interpret AI implementation motivations as either commitment-focused or control-focused, and how these divergent attributions shape policy acceptance in educational settings. Based on an experimental study conducted in Chinese public universities, the findings indicate that greater algorithmic transparency is positively associated with policy acceptance of AI adoption, primarily by reinforcing commitment-focused attributions while mitigating control-focused ones. Notably, organizational reputation exerts an asymmetric moderating influence on the relationship between algorithmic transparency and attributions. Transparency consistently enhances commitment-focused attributions regardless of reputation. However, it only reduces control-focused attributions when organizational reputation is high. These findings highlight the signalling role of transparency in legitimizing algorithmic decision-making in public services and illustrate how organizational reputation shapes the attributional lens through which transparency is interpreted. Taken together, the study provides theoretical insights for research on algorithmic surveillance and practical guidance for policymakers aiming to improve public acceptance in AI adoption.