Dancing with AI: how human-AI interaction affects employee task performance

Abstract

The rapid increase in generative artificial intelligence (Gen AI) products has revolutionised the interaction between humans and artificial intelligence. However, the conceptualisation, measurement, and performance implications of human-AI interaction in the workplace remain underexplored. This study utilised a grounded theory approach to explore the multidimensional construct of human-AI interaction. We then developed a measurement scale for human-AI interaction in the workplace using inductive and deductive methods. Finally, based on self-concept theory, we explored the mechanisms through which human-AI interaction is associated with task performance in the workplace. The results show that (1) human-AI interaction consists of three dimensions: anthropomorphic tool, adaptive trust, and unidirectional emotional connection. (2) The developed scale contained 19 items with good reliability and validity. (3) Human-AI interaction is positively associated with employees’ task performance, and role identity and self-efficacy partially mediate the relationship between human-AI interaction and task performance.