Safety Economics in the AI Era: Global Uncertainty and Crisis Management

Purpose: This study aims to examine the theoretical foundations and practical implications of Safety Economics, focusing on global uncertainty and crisis management in the AI era, particularly whether the Fourth Industrial Revolution can evolve into a genuine “Safety Revolution” integrating technological innovation, economic growth, social safety, and international cooperation. Research design, data, and methodology: The study employs a case study approach analyzing the US-China trade war and semiconductor competition (2018-2025) and the China-Japan trade dispute (1972-2025). The analytical framework applies Keohane and Nye’s three-dimensional concepts of interdependence, sensitivity, and vulnerability. The research examines empirical data on tariff measures, export controls, supply chain disruptions, and incorporates global uncertainty indices, external monetary vulnerability metrics, and economic interdependence indicators. Results: The analysis reveals that AI-era economic risks exhibit nonlinear, multidimensional, and black-boxed characteristics that traditional crisis management cannot address. US-China semiconductor decoupling demonstrates high interdependence coexisting with asymmetric vulnerability, requiring at least $1 trillion in initial investment for complete technological separation. The China-Japan case confirms structural transformation from “political-economic separation” to “political-economic linkage,” showing economic interdependence does not automatically mitigate political conflicts. Risk Management concepts have evolved into multi-layered frameworks integrating enterprise practices with national and global governance. Implications: The findings suggest genuine “Safety Revolution” requires public-private-academic collaborative platforms enabling preventive resource allocation and early intervention. Real-time risk monitoring, supply chain diversification, and digital redundancy are essential. East Asia must establish practical cooperation mechanisms that “freeze” territorial and historical conflicts while introducing confidence-building measures in non-military security domains. Establishing “East Asian Digital Safety Norms” for AI regulation and data governance, and constructing regional crisis management platforms are imperative for sustainable development and human safety.