AI or human steering: unveiling the impact of collaborative systems on employee AI crafting
Mengye Chen, Yeman Cai, Bao Cheng, Yun DongPurpose
Employees' AI crafting plays a critical role in enhancing the effectiveness of human–artificial intelligence (AI) collaboration. Yet, limited research has explored whether different forms of employee–AI cooperation – such as AI-led versus employee-led systems – distinctly shape employees' AI crafting, particularly as these systems convey salient informational cues about organizational expectations and the positioning of employees within the collaborative structure. Drawing on social information processing theory, this study investigates how and when AI-led (vs employee-led) cooperative systems influence employees' AI crafting.
Design/methodology/approach
Two experimental studies (N = 360) were conducted to test our hypotheses. Analysis of variance, ordinary least squares and bootstrapping methods were integrated in the analysis.
Findings
Our results suggest that AI-led (vs employee-led) cooperative systems have a negative indirect effect on employees' AI crafting through their organization-based self-esteem (OBSE). Furthermore, employee knowledge of AI mitigates the negative influence of AI-led (vs employee-led) cooperative systems on employee OBSE, as well as weakens the negative indirect effect on AI crafting.
Originality/value
This research systematically uncovers the potential downside of AI-led (vs employee-led) cooperative systems and offers novel insights into how employee–AI collaboration can be more effectively managed in an increasingly AI-enabled future.