Generative AI as Job Demand and Resource: First Results of an Empirical Study

Nowadays the volume and pace of information is overwhelming, and processing vast amounts of data to make decisions and solve problems requires significant time, effort, and cognitive resources. This can often lead to mental strain for decision-makers. Therefore, to address this challenge, researchers advocate for integrating human and machine intelligence, creating a hybrid system that combines complementary strengths. Generative AI (GenAI) has the ability to process large volumes of information at speeds beyond human capacity and can generate novel ideas to support decision-making. However, decision-makers need training not only to write effective prompts but also to engage in deeper inquiry and reflection. Hence, in this paper, we introduce a new theoretical model, where GenAI is depicted as both a job demand and a job resource. Drawing on the Job Demands-Resources (JD-R) model, we propose that the capacity of a GenAI agent to process large datasets and assist with problem-solving renders it a job resource, while the requirement to acquire new skills to use it effectively constitutes a job demand. Our PLS-SEM analysis supports the hypothesis that GenAI, as a job resource, is positively associated with engagement and negatively associated with burnout in the workplace. However, our findings reveal that GenAI, as a job demand, is not positively associated with burnout; rather, it is positively associated to engagement. Hence, acquiring new skills does not appear to be perceived as a burden by employees.