A study analyzing the pathways through which generative AI affects office workers’ job performance and actual use has been published in the International Journal of Human–Computer Interaction (IJHCI, SSCI), an international journal in the field of human–computer interaction. The joint study was led by first author Jooyoung Kim, a researcher at the Graduate School of Technology and Innovation Management at Hanyang University, with researchers at the HJ Institute of Technology and Management (HJITM) serving as corresponding author.

The researchers surveyed 221 employees at the middle-manager level or above at Korean companies who actively use generative AI in their work, and analyzed the data using partial least squares structural equation modeling (PLS-SEM). They examined whether generative AI’s responsiveness and empathy, along with users’ technology optimism, lead through perceived work performance and practical benefits in customer-related tasks to actual use, and how organizational culture moderates this process.

The analysis showed that generative AI’s empathy significantly increased both perceived work performance (β=0.287) and practical benefits (β=0.257). By contrast, responsiveness — answering quickly and accurately — raised perceived performance (β=0.153) but had no significant relationship with practical benefits. Quick answers alone do not make the tool more useful in practice; responses that take the user’s context into account are what create perceived benefits.

Users’ technology optimism was the factor with the strongest effect on perceived performance (β=0.412) and practical benefits (β=0.296). However, it was perceived performance (β=0.426) that led to actual use, while practical benefits had no direct effect on use. The interpretation is that office workers keep using generative AI less because their work feels easier than because of a sense of accomplishment at doing their job better.

Organizational culture had no direct effect on use by itself, but it strengthened both the link from perceived performance to use (β=0.181) and the link from practical benefits to use (β=0.185). The more open an organization is to innovation and change, the more likely employees who have experienced AI’s effects are to actually use it. Among the control variables, larger company size was associated with lower use (β=−0.182). The researchers suggested that the lengthy approval procedures and complex IT governance of large companies may slow the rapid adoption of new technologies.

The study is significant in explaining the effects of generative AI adoption not in terms of features or speed, but in terms of the quality of interaction represented by empathy and the context of organizational culture. It suggests that to increase generative AI use, companies need not only to roll out the tools but also to create use cases through which employees can experience performance gains and to build an organizational culture that allows new attempts.

HJITM continues joint research with researchers in Korea and abroad on the acceptance and use of generative AI, organizational performance and digital transformation.