Beyond Hybrid Work: Designing High-Performance Organizations in the Age of AI
TL;DR
This paper examines how AI reshapes managerial decision-making by distinguishing decision augmentation from decision automation, and considers the governance tensions between centralized and decentralized approaches to AI deployment, as well as identifying the leadership competencies that gain value once routine managerial tasks are delegated to algorithmic systems.
Abstract
Hybrid work models developed largely in response to workplace disruptions caused by the COVID-19 pandemic, treating location flexibility as the central organizing question. This paper argues that artificial intelligence has moved the debate past questions of where work happens toward questions of how work is structured, decided, and governed. Drawing on Dynamic Capabilities Theory, the Resource-Based View, Knowledge-Based View, Sociotechnical Systems Theory, Organizational Learning Theory, Contingency Theory, and Human Capital Theory, the paper develops a conceptual framework linking AI capability, organizational agility, human-AI collaboration, leadership capability, employee empowerment, organizational learning, innovation performance, and organizational performance. The paper examines how AI reshapes managerial decision-making by distinguishing decision augmentation from decision automation, and considers the governance tensions between centralized and decentralized approaches to AI deployment. It also considers how algorithmic management affects trust, autonomy, and engagement among employees, and identifies the leadership competencies that gain value once routine managerial tasks are delegated to algorithmic systems. Because the paper is conceptual, it proposes a research design using PLS-SEM to test the relationships identified, along with measurement constructs, sampling considerations, and reliability and validity procedures for a future empirical study. The paper closes with implications for managers, policy makers, and organizational designers, and identifies open questions that current literature has not resolved, including the long-term effects of algorithmic management on organizational trust and the conditions under which decentralized AI governance outperforms centralized models. The paper does not report new empirical data. Its contribution lies in integrating separate streams of research on AI adoption, organizational design, and leadership into a single framework intended to guide subsequent empirical testing.