Synthetic knowledge as systemic capital in the governance of generative artificial intelligence

Abstract

Purpose: The diffusion of generative artificial intelligence signals a structural transformation in the architecture of the knowledge economy. While prior research has focused primarily on automation and augmentation (Raisch, Krakowski, 2021; Shrestha et al., 2019), the systemic implications of generative systems for knowledge as capital remain insufficiently theorized. This paper aims to conceptualize synthetic knowledge as systemic capital and to explain its implications for organizational quality, governance and sustainability. Design/methodology/approach: This paper conceptualizes synthetic knowledge as systemic capital, that is, algorithmically generated cognitive content produced within concentrated digital infrastructures and mobilized within organizations. Unlike traditional knowledge capital, held by identifiable members of a firm and integrated through processes the firm controls, synthetic knowledge is externally generated, infrastructurally mediated, and materially constrained. Drawing on the knowledge-based view of the firm (Grant, 1996), socio technical transition theory (Geels, 2002; Markard et al., 2012), and recent scholarship on foundation models and the governance of artificial intelligence (Bommasani et al., 2021, 2024; Burkhardt, Rieder, 2024), the study identifies key structural dynamics of synthetic knowledge production. Findings: The study identifies three structural tensions: cognitive efficiency versus epistemic autonomy, decentralized access versus infrastructural concentration, and digital productivity versus environmental sustainability. Research limitations/implications: This study is conceptual in nature and does not rely on primary empirical data; the framework requires validation across sectors and organizational forms. Practical implications: The findings highlight the need for organizations to develop governance mechanisms for managing synthetic knowledge and epistemic dependencies. Social implications: The study emphasizes the broader societal importance of responsible governance of artificial intelligence and of attention to sustainability. Originality/value: The article develops a conceptual framework linking infrastructural concentration, epistemic asymmetry, organizational quality, environmental externalities, and regulatory governance, including the European Artificial Intelligence Act. It argues that generative artificial intelligence should be understood not as a neutral tool but as a reorganization of cognitive capital embedded in infrastructural and ecological constraints. Keywords: synthetic knowledge; generative artificial intelligence; epistemic asymmetry; organizational quality; governance; sustainability. Category of the paper: Conceptual paper.