Beyond Algorithm Aversion: The Role of Power Relationships in Shaping Mutual Trust in Human-AI Team

In the era of artificial intelligence, users’ attitudes toward algorithms play a critical role in shaping trust and acceptance of AI technologies. However, previous research has overlooked dispositional trust toward algorithms and its role in shaping bidrectional trust in human-AI interaction. To address these gaps, this paper introduces the concept of algorithm aversion tendency (AAT) and develops the AAT scale to capture users’ general trust attitudes towards algorithms. It further examines the cognitive pathways through AAT influences bidirectional trust and behavior intention. The results indicate that AAT has a significant negative effect on trust in AI and behavior intention. The effect remains significant across different human-AI power relationships. However, the effect of AAT on trust is moderated by power relationships. Overall, this study highlights the critical role of dispositional trust in algorithms in shaping trust and behavioral intention in human–AI interaction.