Abstract

Firms continue to invest heavily in artificial intelligence (AI), yet many fail to translate AI capability into measurable organizational value. Existing research has largely treated AI capability as a direct driver of firm performance, offering limited insight into the organizational mechanisms and institutional conditions that explain why firms with similar technological capabilities achieve substantially different outcomes. To address this gap, we ask: how does AI capability translate into firm performance under conditions of institutional resistance? Drawing on the Information Systems (IS) business value literature and institutional theory, this study examines how AI capability translates into firm performance through organizational impact amid institutional resistance. AI capability is conceptualized as an organizational capability that integrates technological infrastructure, analytical resources, and human expertise to improve decision-making and organizational processes (Mikalef & Gupta, 2021). Building on IS business value research, we propose that AI capability influences firm performance indirectly through organizational impacts, including operational efficiency, resource utilization, and process effectiveness (Melville et al., 2004). We further propose that institutional resistance serves as a firm-level boundary condition that constrains AI integration through embedded routines, organizational inertia, and structural barriers. Consequently, firms with similar AI capabilities may experience substantially different performance outcomes depending on their level of institutional resistance. This research contributes to IS literature by shifting attention from direct AI-performance relationships to the organizational mechanisms through which AI creates value. The study further extends IS business value research by incorporating institutional resistance as a structural organizational condition that explains heterogeneous AI outcomes across firms. In practice, we propose a model that provides organizations with a process-based explanation for why firms frequently fail to realize the expected returns from AI investments, even when they develop similar technological capabilities. Additionally, the study offers a theoretically grounded framework that may help managers identify organizational barriers that limit AI-enabled value creation.

Download

Share

COinS