Open Access

ARTICLE

Algorithmic opacity and employees’ knowledge hiding: medication by job insecurity and moderation by employee—AI collaboration

Chunhong Guo¹, Huifang Liu², Jingfu Guo^3,*

1 School of Business Administration, Dongbei University of Finance and Economics, Dalian, 116025, China
2 School of Economics and Management, Shandong Youth University of Political Science, Jinan, 250103, China
3 School of Economics and Management, Dalian Minzu University, Dalian, 116600, China

* Corresponding Author: Jingfu Guo. Email:

Journal of Psychology in Africa 2025, 35(3), 411-418. https://doi.org/10.32604/jpa.2025.065763

Received 20 August 2024; Accepted 19 January 2025; Issue published 31 July 2025

Abstract

We explored the effects of algorithmic opacity on employees’ playing dumb and evasive hiding rather than rationalized hiding. We examined the mediating role of job insecurity and the moderating role of employee-AI collaboration. Participants were 421 full-time employees (female = 46.32%, junior employees = 31.83%) from a variety of organizations and industries that interact with AI. Employees filled out data on algorithm opacity, job insecurity, knowledge hiding, employee-AI collaboration, and control variables. The results of the structural equation modeling indicated that algorithm opacity exacerbated employees’ job insecurity, and job insecurity mediated between algorithm opacity and playing dumb and evasive hiding rather than rationalized hiding. The relationship between algorithmic opacity and playing dumb and evasive hiding was more positive when the level of employee-AI collaboration was higher. These findings suggest that employee-AI collaboration reinforces the indirect relationship between algorithmic opacity and playing dumb and evasive hiding. Our study contributes to research on human and AI collaboration by exploring the dark side of employee-AI collaboration.

---

Keywords

---