AI Governance: A Civilizational Dialogue of the Global South

There are three major models of AI governance in the world today. The United States prioritizes market-driven innovation, technological competition, and national security, pursuing a “small yard, high fence” approach to strategic technologies. The European Union emphasizes legal regulation based on fundamental rights and risk management. China proposes an integration of respect for state sovereignty, inclusive development, technological security, and multilateral cooperation, with the goal of promoting the material and spiritual happiness of peoples.

Supranationalism and multilateralism based on respect for international law are essential to reducing asymmetrical relations among hegemonic forces, constraining the imposition of dominant Global North interests, and building common interests based on respect for the convergence of values among different civilizations. This is also an aspiration of the Global South.

In this context, the design and implementation of AI governance have become a decisive arena for understanding the new dynamics of global power in the 21st century, especially in regulating the growing influence of top technology corporations over policymaking, standards, and digital architectures in the Global South.

Global AI governance is increasingly mired in an institutional fragmentation of different models, reflecting fundamentally different values and assumptions about what governance itself should mean. Therefore, this discussion must recognize the diversity of values as a central explanatory factor. AI acts as an engine of productivity leaps, reshapes the spaces of global interdependence, and influences how societies respond to culturally specific questions of ethics and social norms. Standardized regulatory responses inevitably carry cognitive and normative assumptions, privileging certain values over others. Such imbalances in the alignment of values at the technical level reveal not only institutional deficiencies in existing models of global governance, but also the potential reproduction of a hegemonic vision of the global order that serves the interests of particular segments of the world’s population.

Moving toward more empowering forms of governance requires recognizing that AI is not simply a tool for increasing productivity. It is also about peoples’ ability to preserve their memories, design their institutions, and participate in building technological futures free from subordination. After all, those who have the power to name also have the power to define. In this sense, the Global South cannot simply accept the imposition of foreign technologies. Nations must become active participants in shaping international algorithmic norms, a process that can only be viable if it is grounded in respect for supranationalism and international law.

The cultural-historical approach suggests that the central task is to select, translate and reconfigure technological capabilities according to the needs of each civilization. Therefore, the pluriversality of AI governance does not mean isolation. Rather, it means allowing societies to make their own choices about how technologies are developed and used, while drawing on their own cultural and social contexts.

The transition from the techno-liberal paradigm toward pluriversal governance is not a rhetorical choice, but a question of historical justice. AI can contribute to greater empowerment only if access to knowledge, computing resources, and the ability to shape rules are more widely shared. This raises a fundamental question: if AI is also about who gets to shape the rules and values that govern technology, then AI governance cannot be treated as a purely technical matter. Every regulatory approach reflects particular views of the role of government, markets, communities, individual rights, national sovereignty, and international cooperation.

From this perspective, all countries should adopt a people-centered and inclusive approach and develop AI for the common good, so that AI can become an important driver of shared prosperity and common security. This approach differs from models that place AI primarily in the service of strategic competition or the private concentration of technological capabilities. AI governance should therefore not become a zero-sum competition over models, chips, and data. Instead, it should provide a framework for countries to work together to promote security, development, and effective and legitimate rules for the use of AI.

The author is the Academic Director of the Central American Institute of Public Administration (ICAP).

The article reflects the author’s opinions, and not necessarily the views of China Focus.