AI for Strategic Decision‑Making and Organizational Design

Abstract

This paper examines how artificial intelligence (AI) tools can be integrated into top‑management decision processes and organizational structures without generating harmful dependency, bias, or erosion of human judgment. Drawing on literature from strategic management, organizational design, and human–AI collaboration, I develop theory and hypotheses about three managerial decisions: AI governance structure, use of hybrid human–AI teams, and centralization versus decentralization of AI capabilities. Using simulated cross‑sectional firm‑level data (N = 800 firms across 20 industries) and two‑level multilevel models (firms nested in industries), I illustrate empirical approaches to test the hypotheses. Results from OLS regressions and hierarchical linear models show (simulated) patterns consistent with the theory: stronger internal AI governance (AI governance index) is positively associated with strategic decision quality and reduces the negative association between AI reliance and decision accountability; hybrid teams increase decision speed but their effect on decision quality depends on team design and training; and decentralized AI capabilities improve responsiveness in dynamic environments while centralized capabilities yield higher strategic coherence in stable contexts. I conclude with implications for managers, policy, and future empirical research, provide full R code to reproduce the simulated data and analyses, and discuss limitations of simulation-based inference.