Answers are grounded in this paper and cite their sources.
Purpose: The study aimed to assess the mediating effect of trust and task-technology fit in human-AI collaboration environments in employee engagement in digital workplaces. Design/methodology/approach: A mixed-methods study design was used with 687 employees in various industries. Structural equation modeling (SEM) was used to test the relationships among the capabilities of the AI system, trust, and task-technology fit, as well as relationships between employee engagement and trust. 42 participants in qualitative interviews gave further insights into the interaction patterns between humans and AI. Results: The results show that the capabilities of AI systems substantially contribute to employee engagement via their role of full mediation, namely through their mediation by trust in AI systems (indirect effect = 0.31, p < 0.001) and task-technology fit (indirect effect = 0.27, p < 0.001). The indirect effects of AI Design Quality in the mediation model through trust (0.32) and task-technology fit (0.29) are comparable, resulting in total effects of β = 0.66, p < 0.001. The study shows that design and customization of AI are key factors in boosting employee trust and engagement, and transparency is key to this. Research limitations/implications: Because this was a cross-sectional study, causal inferences are limited. Longitudinal research is needed to investigate the dynamic nature of human-AI interaction in a changing work environment. Practical implications: It is essential for organisations to focus on the transparency of AI, offer thorough training and ensure that tasks and technology align well to maximise the benefits of employee engagement from the implementation of AI. Originality/value: It offers a first comprehensive framework that integrates the Technology Acceptance Model, Job Demands-Resources theory and Human-AI Interaction to gain insights into the impact of AI on employee engagement.
No takes yet. Share an insight, caveat, or question.
Ramkumar et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: