Abstract

As organizations increasingly adopt artificial intelligence (AI) to enhance efficiency and decision-making, a critical question arises: How does AI shape the deeper processes of organizational learning? Drawing on Argyris and Schön’s (1978) distinction between single- and double-loop learning, March’s (1991) exploration-exploitation framework, and Senge’s (1990) systems thinking, this thesis develops a conceptual model that illuminates AI’s potential to both streamline surface-level corrections and catalyze more profound, transformative change.
A literature review spanning human resource development, knowledge management, and organizational behavior reveals that while AI often yields short-term productivity gains, scholars rarely connect these implementations to fundamental learning dynamics such as reflective inquiry, strategic balancing of efficiency and innovation, or system-wide adaptation. Moreover, existing research underscores an array of moderating factors – including leadership style, organizational culture, ethics, and digital maturity – as pivotal in determining whether AI fosters genuine adaptation or merely reinforces existing norms.
By synthesizing classical learning theories with four key AI applications – machine learning/automated decision-making, human-AI collaboration, big data/real-time analytics, and algorithmic feedback – this thesis provides an integrated framework. The model details how AI can facilitate deeper reflection, sustain ambidexterity, and strengthen systemic feedback loops, contingent on contextual moderators and boundary conditions. In so doing, it aims to offer a more holistic lens for scholars and practitioners.