AEMPS

Substitution Boundaries of AI Agents in Team Collaboration: An Empirical Analysis Based on Front-Line Employees' Perceptions

Author

Ziqian Chang * 1

1 Nanjing Foreign Language School

Corresponding Author

Ziqian Chang

Keywords

AI Agent, Human Collaboration, Human-AI Collaboration

Abstract

With AI agents increasingly involved in teamwork, whether AI can substitute human collaboration has become a heated topic in academia. Large-model-driven agents are now capable of participating in team brainstorming, task coordination, and collaborative problem-solving, and knowledge-intensive industries are beginning to form a large number of human-machine hybrid teams. Existing research mostly relies on theoretical reasoning and lacks empirical perceptual data from front-line practitioners. Drawing on a questionnaire survey of 158 employees with cross-industry teamwork experience, this paper explores practitioners' perceptions regarding the substitution of human collaborative behaviors by AI. The results show that AI can only partially substitute procedural and rule-oriented collaborative tasks, whereas social-emotional communication, conflict mediation and value judgment remain irreplaceable human strengths. Position, work experience and AI familiarity jointly shape individual perceptions of the substitution boundary. Most practitioners regard AI as an auxiliary tool rather than a leader of team collaboration. This study provides a practical reference for enterprises applying AI collaboration tools.

Citation

Ziqian Chang. Substitution Boundaries of AI Agents in Team Collaboration: An Empirical Analysis Based on Front-Line Employees' Perceptions. AEMPS (2026) Vol. 315: 69-74. DOI: 10.54254/2754-1169/2026.38925.

References

[1]: Su, J. (2026). Can AI replace human work? - On the new round of economic mediation driven by intelligent technology.Journal of Shandong Normal University (Social Science Edition), 71(1), 147-159.

[2]: Chen, H., & Feng, C. (2026). When employees meet AI: Concept measurement, pre-cause configuration and impact mechanism of employee-AI collaboration.Progress in Psychological Science, 34(8), 1309-1329.

[3]: Du, H., & Ding, Z. (2026). Research on the impact of organizational AI transparency on the satisfaction of human-computer collaboration.China Human Resources Development, 43(5), 6-23.

[4]: Jarrahi, M. H., Askay, D., Eshraghi, A., & Smith, P. (2023). Artificial intelligence and knowledge management: A partnership between human and AI.Business Horizons, 66(1), 87-99.

[5]: De Cremer, D., & Kasparov, G. (2021). AI should augment human intelligence, not replace it.Harvard Business Review, 18(1), 1-8.

[6]: Li, C., Zhang, Y., Niu, X., Chen, F., & Zhou, H. (2023). Does artificial intelligence promote or inhibit on-the-job learning? Human reactions to AI at work.Systems, 11(3), 114.

[7]: Vaccaro, M., Almaatouq, A., & Malone, T. (2024). When combinations of humans and AI are useful: A systematic review and meta-analysis.Nature Human Behaviour, 8(12), 2293–2303.

[8]: Peng Zhen, Kang Siwei, Zhang Jilong & Feng Ping. (2026). Man-machine dancing or human competition? Work reshaping after AI intervention and its cross-level impact on the psychological experience of knowledge workers.China Human Resources Development, 43(6), 38–58.

[9]: Wang, D., Churchill, E., Maes, P., Fan, X., Shneiderman, B., Shi, Y., & Wang, Q. (2020). From human-human collaboration to Human-AI collaboration: Designing AI systems that can work together with people. InExtended abstracts of the 2020 CHI conference on human factors in computing systems(pp. 1–6).

[10]: Cabrera, A. A., Perer, A., & Hong, J. I. (2023). Improving human-AI collaboration with descriptions of AI behavior.Proceedings of the ACM on Human-Computer Interaction, 7(CSCW1), 1–21.

Copyright © 2021 Eliwise Academy. Unless Otherwise Stated