员工-人工智能合作如何影响领导感知员工努力？

金杨华^1,2　朱汕怡¹　吴珍珍¹　谢江佩¹

（1.浙江工商大学；2.浙江金融职业学院）

摘要：在人工智能（AI）广泛应用于组织的背景下，员工与AI合作日益普遍，但对于领导如何解读员工此类行为并形成相应的管理判断尚缺乏较为系统的理论探讨和实践检验。本文通过两项研究探讨了员工-AI合作影响领导感知员工努力的机制及其管理后果。情境实验发现，员工-AI合作对领导感知努力的影响受到领导印象管理归因及领导AI信任的调节：当领导将员工行为归因为印象管理的倾向更低时，员工-AI合作更易被感知为努力；当领导AI信任程度更高时，员工-AI合作更容易被领导感知为努力。领导-下属配对问卷调查进一步验证了其中的调节效应：领导印象管理归因通过调节员工-AI合作与领导感知员工努力的关系，间接影响员工工作绩效和领导授权；领导AI信任通过调节员工-AI合作与领导感知员工努力的关系，间接影响员工工作绩效和领导授权。本文揭示了AI时代领导者社会认知的内在机制，为推动人与AI协作的组织管理实践提供了理论参考。

关键词：员工-AI合作；领导感知员工努力；印象管理归因；AI信任

作者简介：金杨华，浙江工商大学工商管理学院教授、博士生导师，杭州，310018；浙江金融职业学院教授，杭州，310018；朱汕怡，浙江工商大学工商管理学院博士研究生；吴珍珍，浙江工商大学工商管理学院硕士研究生；谢江佩，浙江工商大学工商管理学院副教授，通信作者。

基金项目：国家社会科学基金重点项目“人工智能算法劳动的权力平衡研究”（22AGL014）

引用格式：金杨华，朱汕怡，吴珍珍，等. 员工-人工智能合作如何影响领导感知员工努力？［J］. 经济与管理研究，2026，47（9）：146-160.

How Does Employee-AI Collaboration Affect Leaders' Perceptions of Employee Effort?

JIN Yanghua^1,2, ZHU Shanyi¹, WU Zhenzhen¹, XIE Jiangpei¹

(1. Zhejiang Gongshang University, Hangzhou 310018;

2. Zhejiang Financial College, Hangzhou 310018)

Abstract: Artificial intelligence (AI) integration into organizational workflows has transformed employee-AI collaboration into a routine work practice. Existing research focuses primarily on the performance benefits and psychological outcomes of AI adoption from the employee perspective. However, less attention is paid to how leaders, as key observers, interpret this collaboration and form managerial judgments. This paper investigates when and why leaders perceive employee-AI collaboration as genuine effort, and how this perception shapes performance evaluations and empowerment decisions.

Attribution theory indicates that employee-AI collaboration carries no fixed meaning of effort. AI reduces the visibility of traditional effort cues while providing a strong external explanation for outputs, requiring leaders to decode the motivational meaning of employee behavior. Two leader-side cognitive frameworks govern this decoding process: impression-management attribution reflects the extent to which leaders attribute the collaboration to impression-enhancing motives. Trust in AI reflects the extent to which leaders view AI as a capable and reliable work resource. These frameworks shape leaders' perceptions of employee effort, which then serves as a proximal mechanism linking employee-AI collaboration to performance evaluation and empowerment.

Two complementary studies tested the proposed model.Study 1 employed a scenario-based experiment among practicing managers to test the moderating roles of impression-management attribution and trust in AI. Study 2 used a multi-wave, multi-source field survey of leader-employee dyads to replicate these interactive effects and examine downstream managerial consequences. Results supported the moderated mediation hypotheses: the indirect effects were more positive under low impression-management attribution and less negative under high trust in AI.

This paper makes three theoretical contributions. First, it advances third-party evaluations of AI-related behavior from distal outcomes such as performance inflation or laziness attributions to leaders' perceptions of employee effort as a critical proximal cognitive process. Second, it develops an integrative dual-pathway framework that incorporates both motivational cues of impression-management attribution and instrumental cues of trust in AI, thereby extending attribution theory to the emerging context of employee-AI collaboration. Third, by simultaneously examining performance evaluation and empowerment, it demonstrates that effort perceptions influence not only the recognition of employees' past contributions but also the allocation of future autonomy and developmental opportunities. Practically, the findings suggest that organizations should help leaders update traditional effort metrics and cultivate greater trust in AI technologies, so as to foster more accurate recognition of employee contributions in employee-AI collaboration.

Keywords: employee-AI collaboration; leaders' perceptions of employee effort; impression-management attribution; trust in AI

下载全文