THE IMPACT OF GENERATIVE AI ADOPTION ON BUSINESS PERFORMANCE: THE MEDIATING ROLE OF EMPLOYEE PRODUCTIVITY AND THE MODERATING ROLE OF DIGITAL LITERACY

Abstract

This study examines the relationship between Generative Artificial Intelligence (GenAI) adoption and business performance among organizations in Malaysia, with a specific focus on the mediating role of employee productivity and the moderating role of digital literacy. Anchored in the Resource-Based View (RBV) and the Technology-Organization-Environment (TOE) framework, a conceptual model was developed and tested using survey data collected from 345 managers and business owners across key Malaysian economic sectors. Data were analysed using partial least squares structural equation modeling (PLS-SEM) in SmartPLS 4.0.The results indicate that GenAI adoption has a significant positive direct effect on business performance and a stronger positive effect on employee productivity. Employee productivity was found to significantly and partially mediate the relationship between GenAI adoption and business performance. Digital literacy was found to significantly moderate the GenAI adoption-employee productivity relationship, such that the positive effect of GenAI adoption on productivity is considerably stronger among firms with digitally literate workforces. A moderated-mediation analysis further confirmed that the indirect effect of GenAI adoption on business performance through employee productivity is conditional on the level of digital literacy. The findings offer practical guidance for Malaysian business leaders and policymakers under the MyDIGITAL initiative and the National AI Roadmap, suggesting that the performance returns of GenAI investment are not automatic but are substantially contingent on complementary investment in workforce digital capability. This study contributes one of the first empirically framed investigations of GenAI's productivity-performance pathway in the Malaysian context, and is among the first to formally test digital literacy as a boundary condition of this relationship using a moderated-mediation design.