Dynamic Global Alignment Model (DGAM): An AI-Driven Framework for Computational Geopolitics and India's Strategic Statecraft in the 21st Century

This research introduces the Dynamic Global Alignment Model (DGAM), a pioneering AI-driven framework that redefines the study of computational geopolitics and strategic statecraft in the 21st century. By synthesizing reinforcement learning (MDP formalism), Bayesian inference, and scenario forecasting, DGAM advances a rigorous, data-driven methodology for understanding power shifts, risk landscapes, and alignment dynamics in international relations. The thesis focuses on India’s evolving role in global politics, situating its economic resilience, trade connectivity, and defense modernization within the broader context of multipolarity and AI governance. Using authoritative datasets (IMF, World Bank, WITS/UN Comtrade, SIPRI) and robust computational modeling, the study provides transparent, reproducible, and policy-relevant insights into India’s strategic options through 2025–2035. Key Contributions: • Formalization of DGAM as a multi-agent Markov decision process for global strategy analysis. • Integration of international relations theory with cutting-edge AI and computational social science. • Comparative evaluation of DGAM against baseline IR models, including ablation and robustness studies. • Scenario-based forecasting of India’s strategic risks and opportunities in a volatile global order. • Comprehensive appendices with reproducibility checklists, hyperparameters, and open data transparency. This work represents a rare intersection of theoretical innovation and applied policy relevance, designed for scholars, policymakers, and technologists seeking to navigate the complexities of geopolitics in the age of artificial intelligence.