Researchers Map AI Multilateralism’s Principles and State Relations

The increasing prominence of artificial intelligence presents a significant challenge to global governance, demanding new frameworks for international cooperation. Michal Natorski of Maastricht University/UNU-MERIT, alongside Markus Furendal and Magnus Lundgren from Edward Elgar Publishing Ltd, investigate how multilateralism attempts to address this rapidly evolving landscape. Their work analyses the core principles guiding international efforts in AI governance and examines how states coordinate their responses to this technology. The research reveals that while AI governance involves multiple stakeholders, states maintain a dominant role in shaping agendas and implementing international agreements, highlighting a persistent hierarchy within this new sphere of cooperation.

Fragmented AI Governance and Growing Urgency

The global landscape of artificial intelligence governance is rapidly evolving, presenting both immense opportunities and significant risks. This research demonstrates a growing urgency for international cooperation, yet achieving it proves difficult due to diverging national interests, differing values, and the sheer speed of technological development. The study highlights a shift from initial discussions to concrete, though often contested, actions within the UN system and other international forums. There is increasing recognition of AI’s potential to transform economies and societies, necessitating proactive governance rather than simply reacting to developments.

AI is framed as a general-purpose technology, comparable to electricity or the internet, with the potential to reshape all sectors, underscoring the need for careful planning and international collaboration. The research also explores how our understanding of AI’s future shapes its development and governance, as different visions lead to different priorities. A recurring theme is the tension between national sovereignty and the need for international cooperation, as countries are reluctant to relinquish control while recognizing the benefits of global standards. Capacity building is crucial, particularly for developing countries, to ensure they can benefit from AI and participate meaningfully in global governance discussions.

The research consistently frames AI as presenting both risks, such as bias, job displacement, and security threats, and opportunities, including economic growth, improved healthcare, and solutions to climate change. The United Nations is presented as the central forum for global AI governance, with various bodies playing a role, and the UN AI Advisory Body’s report is a key contribution. Several countries, including the US, China, the Netherlands, Austria, and the Republic of Korea, are actively involved in shaping the agenda. The Global Partnership on AI is identified as an important multi-stakeholder initiative, though its limitations, such as a lack of binding authority, are also acknowledged.

UNESCO plays a role in setting ethical standards, while China is increasingly assertive in promoting its own vision of AI governance. The research highlights several challenges to effective global governance, including a lack of consensus on key issues like data governance and algorithmic bias, the rapid pace of technological change, and difficulties in enforcing international agreements. Power imbalances and geopolitical competition further complicate the landscape. Recent UN activities, such as resolutions on lethal autonomous weapons systems and initiatives focused on AI for sustainable development, demonstrate growing engagement.

The Pact for the Future, including the Global Digital Compact, aims to bridge the digital divide and promote responsible technology use. The Security Council is also increasingly discussing the security implications of AI, utilizing informal meetings and expert briefings to inform its deliberations. Potential pathways forward include strengthening multilateralism, developing common standards and norms, promoting capacity building, fostering multi-stakeholder dialogue, adopting a risk-based approach, and promoting transparency and accountability. In conclusion, the research paints a complex picture of global AI governance. While there is growing recognition of the need for international cooperation, achieving it will be challenging due to diverging national interests, rapid technological change, and the lack of a strong enforcement mechanism. The UN is playing a central role, but its effectiveness will depend on the willingness of member states to compromise and work together.

Discourse Analysis of AI Multilateralism Principles

This research investigates how international cooperation, or multilateralism, is responding to the emergence of artificial intelligence as a transformative technology. The study adopts a unique approach by examining not just what is being discussed in international forums, but how the very language and underlying principles of those discussions shape the responses to AI. Rather than simply cataloging existing initiatives, the research focuses on the core beliefs and assumptions that frame the debate around AI governance. The methodology centres on a detailed analysis of the discourse surrounding AI multilateralism, identifying key principles that legitimize particular actions by states.

Researchers examined how concepts of epochal change, technological determinism, and dialectical understanding are used to justify approaches to AI governance, and how these principles influence the integration of AI into existing international frameworks. This involved a close reading of publicly available information, including reports from organizations like the United Nations, and direct observation of debates at events such as the AI for Good Summits and Responsible Artificial Intelligence in the Military Domain summits. A key aspect of the research is its emphasis on understanding the role of states within this evolving landscape. While acknowledging the increasing involvement of non-state actors, the study demonstrates that states consistently maintain a dominant position in setting agendas, negotiating agreements, and implementing international commitments.

This hierarchical dynamic is explored through analysis of how states frame discussions and exert influence within multilateral forums. The research does not rely on technical assessments of AI capabilities, but instead focuses on the political and diplomatic processes shaping its governance. This innovative approach lies in its combination of discourse analysis and observation of international forums, allowing researchers to trace the evolution of AI multilateralism and identify the underlying power dynamics at play. By examining the principles that legitimize state actions, the research provides a nuanced understanding of how international cooperation is adapting to this transformative technology, and what this means for the future of global governance. The findings contribute to a broader understanding of how multilateralism functions in the face of technological disruption, and how states navigate the challenges and opportunities presented by general-purpose technologies like artificial intelligence.

States Dominate Global AI Governance Efforts

The emerging landscape of artificial intelligence governance is characterized by a complex interplay between nations and other stakeholders, revealing a distinctly hierarchical structure despite efforts at broad collaboration. Research demonstrates that while many actors participate in discussions surrounding AI’s development and deployment, states consistently retain ultimate authority in shaping AI global governance. This is not to say that non-state actors are insignificant; their presence influences the process, yet states remain the decisive decision-makers. Current initiatives, such as the UN General Assembly resolution on AI and regional strategies developed by organizations like the African Union and ASEAN, illustrate a growing recognition of the need for international cooperation.

However, these efforts often coexist with more restricted, state-led multilateralism, suggesting a fragmented approach to global governance. While the United States and China have selectively promoted multilateral discussions, their willingness to collaborate is tempered by the sovereignist positions of nations like Russia and Iran, and the general indifference of many other countries. This results in a system where tangible results are limited, and the UN framework often serves as an additional layer alongside these more exclusive arrangements. Despite this complexity, a clear pattern emerges: states translate broad, non-legally binding international commitments into enforceable laws within their own jurisdictions.

This highlights the crucial connection between global and regional regulations, and between state and non-state actors. The development of norms and rules defining appropriate AI usage is progressing, but their effectiveness relies on states’ willingness to integrate them into national legal frameworks. Recent initiatives, like the EU AI Act, demonstrate this process, showcasing how regional approaches can influence global standards. Furthermore, research indicates that the proliferation of AI is prompting a re-.