PUBLISHED
May 31, 2026

OpenAI has recently put forward an idea that could reshape how the world thinks about artificial intelligence governance. It marks a turning point. The conversation is no longer theoretical — it is becoming institutional. The proposal, according to Bloomberg, envisions a multilateral global body, led by the US but also including China as co-equal stakeholder, to coordinate safety standards, oversight mechanisms, and risk management for advanced AI systems.

OpenAI’s idea comes at a critical moment. AI is no longer just a technological breakthrough; it is rapidly transforming economic and geopolitical competition. Without a unified and inclusive system of governance, the world risks repeating past mistakes — where powerful technologies, driven by profit rather than responsibility, created widespread harm. Social media offers a cautionary tale. AI could be far more consequential.

The proposal reflects a growing realisation within the industry: fragmented national regulations are no longer enough. AI systems are evolving too quickly, and they operate across borders too seamlessly. What is needed instead is structured global coordination — an approach that matches the scale and speed of the technology itself.

OpenAI’s position signals a shift. The company is moving beyond neutrality and stepping into active policy advocacy. This is significant. It suggests that even leading developers now recognise that AI has outgrown traditional regulatory models. No single country, no matter how powerful, can manage its risks alone.

At the heart of the proposal lies a delicate balancing act. While the framework would place the US in a leadership role, it also calls for China to participate as an equal stakeholder. This dual approach acknowledges two realities: American dominance in AI innovation, and China’s parallel technological rise. Excluding either would undermine the system.

China’s inclusion is particularly strategic. It aims to prevent the world from splitting into competing AI governance blocs — one Western-led, the other China-centric. Such a divide could deepen mistrust, reduce transparency, and ultimately increase global risk. Cooperation, even among rivals, becomes not just desirable but necessary.

The urgency behind the proposal is driven by growing global anxiety. Policymakers worldwide are increasingly concerned about the risks posed by advanced AI. These risks are varied and serious. They include unpredictable system behaviour, misuse by malicious actors, military escalation, and declining transparency as models become more complex and proprietary.

What was once hypothetical is now immediate. AI systems are becoming capable of reasoning, planning, and interacting with real-world environments. Governance gaps are no longer future problems — they are present vulnerabilities. At the same time, the broader industry is undergoing a shift in mindset. For years, major technology companies resisted strict regulation, fearing it would stifle innovation. That resistance is fading. Today, many leading developers are calling for coordinated global oversight. They argue that inconsistent rules across countries create loopholes, weaken enforcement, and distort competition.

In this new context, safety standards are no longer seen as barriers. They are becoming the foundation for responsible innovation. Without them, scaling AI across borders could prove dangerous.

Although the proposal is still in its early stages, its intended functions are becoming clear. It would establish global safety benchmarks, enabling consistent evaluation of AI systems. It would facilitate information sharing between nations, helping identify emerging risks. It would coordinate policies for high-risk applications, particularly in areas like defense and critical infrastructure. And perhaps most importantly, it would create a permanent channel for dialogue between the US and China —even during periods of political tension. In essence, the goal is simple but ambitious: to institutionalise communication in a domain where competition is intensifying, yet cooperation remains essential.

This proposal does not stand alone. It aligns with broader international efforts, including the Global AI Governance Action Plan released during the 2025 World Artificial Intelligence Conference and High-Level Meeting on Global AI Governance.

That framework describes AI as both a “new frontier in human development” and an “international public good,” while simultaneously warning of unprecedented risks and challenges. It calls for global cooperation across multiple domains, including infrastructure development, data governance, safety standards, and industrial integration.

Key themes include: treating AI as a shared development opportunity; promoting cross-border innovation and research cooperation; expanding digital infrastructure, especially in developing countries; establishing global safety and ethical standards; and strengthening multilateral coordination under UN-linked mechanisms.

Taken together, these initiatives reflect an emerging consensus that AI governance must be global in scope, even if implementation remains fragmented. AI governance is inseparable from geopolitics. The US and China are locked in a high-stakes technological rivalry — one that spans semiconductors, cloud systems, and military applications. AI sits at the centre of this contest.

This creates a paradox. Competition drives innovation, but it also increases the risk of fragmentation and escalation. The OpenAI-proposed governance model attempts to manage this tension. It does not eliminate rivalry; instead, it seeks to contain it within agreed rules and communication frameworks. Analysts often describe this as “managed competition.”

History offers some parallels. In nuclear governance, rival states competed fiercely while still cooperating through treaties and monitoring systems. But AI is different. It is more decentralised, more accessible, and evolving at a far faster pace. This makes regulation significantly more complex.

The challenges ahead are substantial. Deep mistrust between major powers could limit cooperation. Enforcement mechanisms may prove weak or inconsistent. A handful of private companies control much of AI development, complicating state-led governance. And above all, the technology itself is advancing faster than institutions can adapt.

For these reasons, any global governance system is unlikely to emerge overnight. It will take shape gradually, through negotiation, experimentation, and incremental agreements.

The question is no longer whether AI should be regulated, but how the world will organise that regulation in an era defined by both innovation and rivalry.

Ultimately, artificial intelligence is more than a technological issue. It is a structural force—one that is reshaping global power, economic systems, and international order. The challenge ahead is to balance three competing imperatives: innovation, security, and geopolitical competition. How the world manages this balance will not only determine the future of AI. It may also shape the stability of the global system for decades to come.