The Effect of Source Disclosure on Trust in an AI-Based Knowledge Management System

Knowledge management systems (KMS) based on artificial intelligence (AI) can improve retrieval and exchange of human-generated knowledge in organizations. When interacting with an AI-based KMS, the disclosure of the human source could increase system transparency and influence cognitive trust formation. In a vignette-based experiment with 590 employees, we analyze how source disclosure, experimentally varied by familiarity and credibility, affects trust in AI-based KMS. Furthermore, we examine the impact of employees’ risk affinity, people-pleasing propensity, and perfectionism on their intention to use AI, depending on the criticality of the decision situations. Findings show that source disclosure does not lead to higher overall trust but enables trust adjustment according to perceived source credibility. Personality traits and criticality of the situation further influence users’ intention to interact with the system. The results contribute to human-computer interaction research on system transparency, trust calibration, and human factors, offering implications for designing trustworthy, human-centered AI.