人工智能自主性对用户感知和消费行为的影响与作用机制——基于元分析的方法

The Impact and Mechanism of Artificial Intelligence Autonomy on User Perception and Consumption Behavior—Method Based on Meta-Analysis

摘要: 当前众多研究者对用户接受人工智能技术的兴趣和研究激增。然而，现有的研究显得分散，缺乏系统的综合，限制了用户对人工智能技术接受度的理解。文章从用户感知价值角度结构人工智能产品用户感知利益与感知风险权衡，及从用户感知技术视角解释人工智能产品有用性和易用性效应机制，对现有研究中各影响因素进行研究，探索人工智能的自主技术对消费者的影响。为有效帮助预测人工智能产品自主性方面发展，对企业如何设计和消费者使用人工智能提供参考与建议。文章运用元分析方法，对42篇实证研究文献212个效应值进行归纳，得出8个影响因素与用户对人工智能产品购买意愿行为有相关关系，其中感知利益的相关性最强，拟人化的相关性最弱。同时，更高水平感知有用性、感知易用性、社会参与、信任会增强用户购买意愿，更高水平感知风险会降低用户购买意愿。人工智能类型及应用领域在感知机制中起调节作用。文章厘清了人工智能产品自主性对用户购买意愿的效应机制，解构了用户感知价值和技术感知的内在理论逻辑，进一步拓展探索后续研究影响因素效应边界。

Abstract: Currently, there is a surge in interest and research among numerous researchers regarding users’ acceptance of artificial intelligence technology. However, existing research appears scattered and lacks systematic integration, which limits users’ understanding of the acceptance of artificial intelligence technology. The article constructs the balance between perceived benefits and perceived risks of artificial intelligence products from the perspective of user perceived value, and explains the effectiveness mechanism of AI product usefulness and usability from the perspective of user perceived technology. It studies various influencing factors in existing research and explores the impact of AI’s autonomous technology on consumers. To effectively assist in predicting the autonomous development of artificial intelligence products and provide references and suggestions for how enterprises design and consumers use artificial intelligence. The article uses meta-analysis to summarize 212 effect values from 42 empirical research papers, and identifies 8 influencing factors that are related to users’ willingness to purchase artificial intelligence products. Among them, the correlation between perceived benefits is the strongest, and the correlation between personification is the weakest. At the same time, higher levels of perceived usefulness, perceived ease of use, social participation, and trust will enhance users’ willingness to purchase, while higher levels of perceived risk will reduce users’ willingness to purchase. The types and application areas of artificial intelligence play a regulatory role in perception mechanisms. The article clarifies the mechanism of the effect of artificial intelligence product autonomy on user purchase intention, deconstructs the inherent theoretical logic of user perceived value and technology perception, and further expands the exploration of the boundary of influencing factors in subsequent research.