Artificial intelligence is no longer discussed only by engineers and technology companies. It now affects economies, security, human rights, public services, and societies worldwide. At the same time, governments are developing their own rules and policies for dealing with AI. This raises a difficult international question: How can countries govern a technology that transcends national borders when their approaches often differ? This tension between reducing direct government control and relying on governments to establish effective rules reflects a broader political paradox.
The United Nations is trying to address this issue through its Global Dialogue on AI Governance. The initiative is not intended to replace national AI laws with a single global system. Instead, it gives countries and other stakeholders a place to discuss their approaches and identify areas for cooperation. In this article, I examine how this process developed, what the first dialogue revealed, and what the UN can realistically achieve.
Why Has AI Governance Become a Global Issue?
AI is not limited to a single country. It relies on global systems that cross national borders, including computer hardware, cloud platforms, data, AI models, applications, and digital services. Large technology companies also control much of the data, computing power, advanced chips, and AI models used worldwide. According to the UN paper, AI can affect other countries through misinformation, cyberattacks, autonomous weapons, digital services, and shifts in global markets. As competition in AI grows, decisions made in one country can have consequences far beyond its borders. For this reason, countries should manage AI’s risks and benefits through international cooperation and shared responsibility rather than acting alone.
International coordination is important because AI governance is developing at different rates around the world. Countries have different laws, resources, and levels of technical expertise. The UN paper explains that governments are already using different rules, principles, and tools to manage AI. Without cooperation, these approaches could become inconsistent and fragmented. The UN’s role is to bring together governments, technology companies, researchers, and civil society. By working together, these groups can develop common standards, share information, build consensus, and help countries strengthen their ability to govern AI. Effective global AI governance should also include research, coordination, common standards, training, monitoring, and accountability. Simply put, countries need to cooperate because some AI-related challenges are too large for any one country to solve alone.
AI should be governed in accordance with international law, including international human rights law. It can affect privacy, freedom of expression, equality, and fair treatment, so these rights must be protected at every stage of AI’s development and use. AI can also threaten peace and security through autonomous weapons, misinformation, cyberattacks, and other harmful applications. At the same time, it can support innovation, improve decision-making, and help countries achieve the Sustainable Development Goals. However, these benefits are not shared equally. AI governance should support innovation while protecting social, economic, and environmental interests. It is not only about controlling dangerous technology; it is also about ensuring that AI protects people and supports sustainable development.
Another major issue discussed in the paper is the widening AI gap between countries. The AI industry depends on hardware, advanced chips, cloud computing, data, models, and skilled workers. However, the companies that control many of these resources are based in only a few countries. Developing countries may therefore be left with lower-paid work, such as data labelling and content moderation, while receiving fewer economic benefits. The message is clear: AI could widen the gap between developed and developing countries. Many developing countries also have limited access to computing power, cloud services, reliable energy, data, and technical expertise. To address this problem, the UN calls for greater investment in skills, technology, data, and computing resources. It also supports stronger technical and regulatory training for policymakers and AI professionals, especially in developing countries.
Why Is International Cooperation Necessary for AI Governance?
One point I would like to make clear is that different regions have different AI policies. This makes it difficult to present one unified approach. For example, the AIPRM report covers more than 70 countries, as well as the European Union, and shows the different AI policies and initiatives developed around the world. The 10 countries with the most AI policies also have their own organizations and strategies. It is still unclear how all these different interests and strategies can work together on one platform. Take the European Union’s AI Act as an example. Here are five of its most important rules:
1. AI is divided into four risk levels: unacceptable risk, high risk, limited risk, and minimal risk.
2. High-risk AI is strictly regulated: most obligations apply to companies that develop and provide high-risk AI systems.
3. AI users also have responsibilities: organizations and public authorities that use high-risk AI systems must follow certain rules.
4. General-purpose AI has specific requirements: providers must share technical information, follow copyright rules, and publish a summary of the content used for training.
5. Systemic-risk AI faces stronger rules: providers must carry out evaluations, manage risks, report serious incidents, and maintain cybersecurity.
These differences between national approaches show why international cooperation is important. The UN gives countries a platform to discuss their approaches and explore ways to make them work better together.
How Did the UN Establish the Global Dialogue?
The idea for the UN’s Global Dialogue on AI Governance came from the Global Digital Compact, which was adopted in September 2024 as part of the Pact for the Future. The Compact called for stronger international cooperation on AI and encouraged countries to make their different approaches to AI governance more compatible. It also proposed a Global Dialogue where governments and other stakeholders could discuss these issues. In August 2025, the UN General Assembly officially established the Global Dialogue through Resolution A/RES/79/325. Its first annual meeting was held in Geneva on 6–7 July 2026. More than 3,000 people attended, including representatives from 163 countries.
What Did the First Dialogue Reveal?
The first Dialogue showed that many countries are dealing with similar questions about AI, even though their laws and policies are different. The discussions covered the benefits and risks of AI, unequal access to computing resources and skills, the need for safe and trustworthy systems, and concerns about human rights and accountability. The Co-Chairs described these as areas where common views were beginning to emerge, but they made it clear that the summary was not a negotiated agreement. This distinction is important. The meeting did not create a single global AI law. Instead, it gave countries and other stakeholders a place to compare their concerns, experiences, and approaches. The UN sees the Dialogue as an ongoing process, with the next annual meeting scheduled for May 2027.
What Can the UN Actually Achieve?
The UN is unlikely to make every country adopt the same AI laws. A more realistic role is to help countries understand one another’s approaches, find areas of common ground, share useful practices, and make their governance systems work better together. The Dialogue is not a negotiating forum and does not produce binding agreements. Its influence will therefore depend on whether these discussions lead to practical cooperation beyond the meetings. Rather than acting as a single global regulator, the UN can provide a space where governments, researchers, companies, and civil society work toward a shared understanding. The real test will be whether this process leads to cooperation that countries can use in practice.
AI is developing across borders, while governments are creating different ways to manage it. This makes international coordination increasingly important. The UN’s Global Dialogue brings these different approaches into the same conversation without requiring every country to follow an identical legal system. The first meeting did not solve the challenges of global AI governance, but it created an ongoing process for discussion and cooperation. The next question is whether these conversations can lead to more compatible approaches and give countries with fewer AI resources a meaningful role in shaping the rules of the emerging AI world.