Summary of Key Points
In the summer of 2026, seven AI governance summits were held around the world (including the G7, the United Nations, and China's WAIC/APEC Digital Week), signaling a shift in focus from AI technology competition to governance system competition. Both China and the United States have established their own cooperation platforms (the U.S.'s Pax Silica and China's WAICO), leading to a dual-track approach. At the same time, a new trend of co-governance by governments and enterprises has emerged, with an increasing voice for the global South (developing countries). Market ecosystems are beginning to split due to governance differences, and the global AI governance framework is in the process of being restructured.
Detailed Analysis
1. China and the U.S. Establishing Their Own AI Governance Platforms
China and the U.S. have created two distinct AI governance platforms with vastly different objectives:
- Pax Silica (Silicon-based Peace Mechanism): This platform focuses on ensuring the security of critical supply chains, including semiconductors, AI, and rare earths. Its partners are mainly developed and emerging economies in Europe, America, and Latin America (with new additions such as Germany, the Netherlands, and Argentina), indicating a somewhat exclusive nature. The U.S. has also allocated funds from the Chip Act to establish the "Silicon-based Peace Fund" to pilot full-chain traceability at ports in Panama, aiming to prevent supply disruptions.
- WAICO (World Artificial Intelligence Cooperation Organization): This platform aims to promote "inclusive AI" by making AI technology accessible to developing countries and narrowing the technological gap. Its signatories are primarily from the global South, with the headquarters located in Shanghai. China has released a collection of AI application cases from over 20 countries and cooperation plans in eight areas, such as intelligent computing power and open-source ecosystems.
The difference between the two platforms is clear: the U.S. seeks to secure its allies' interests, while China aims to facilitate the development of more countries.
2. Diverse Approaches on Multilateral Platforms
The main global multilateral organizations also have their own roles in AI governance:
- United Nations Global AI Dialogue: The most inclusive platform, with participation from 193 countries, as well as businesses and academia. However, there are significant disagreements, with governments focusing on ensuring that developing countries can benefit from AI, while businesses and academia are concerned about AI security risks. The UN Secretary-General has called for regulated development and proposed initiatives on AI safety for children and renewable energy data centers.
- G7 Évian Summit: A Western elite forum that for the first time brought together executives and leaders from AI giants like OpenAI and Anthropic to discuss AI supply chain security and digital privacy for minors, reflecting a Western-centric approach to rule-making.
- APEC Digital Week: Hosted by China, this event focused on AI innovation, data circulation, and industrial transformation in the Asia-Pacific region, with ministerial meetings and public workshops on AI literacy. China used this platform to promote its own governance concepts in the region.
These differences have created three distinct layers of AI governance: the inclusive UN, the Western-led G7, and the Asia-Pacific-focused APEC.
3. Governance Divisions Impacting the Market
The rivalry between China and the U.S. is affecting businesses and consumers:
- Companies Turning to Chinese Models: Due to U.S. restrictions on the export of advanced AI models and higher costs, many global companies are adopting Chinese models or combining Chinese and American technologies.
- Western Public Concerns: There is anxiety in the West about AI displacing jobs, but there is a lack of awareness about how AI can also enhance industries and improve lives (e.g., in healthcare and agriculture).
- Vicious Cycle: Geopolitical rivalries in AI governance platforms lead to corporate choices of ecosystems, which in turn exacerbate market divisions, reinforcing the bipolar landscape.
4. Co-governance Becoming the New Norm
Enterprises are no longer passive recipients of governance; they are now active participants:
- G7 Summits: AI executives are invited to discuss governance rules directly with leaders.
- ITU AI Committee: The committee includes representatives from Rwanda, Salesforce, and NVIDIA, demonstrating a collaborative approach to governance.
In summary, AI governance is becoming more fragmented, with multiple stakeholders (governments, businesses, and international organizations) playing distinct roles.
5. Five Trends for the Future of AI Governance
Based on the summits, five new trends are emerging:
- Collaboration between Governments and Enterprises: Enterprises are moving from being regulated to co-creating governance rules.
- Multilateral Leadership: Inclusive platforms like the UN counteract Western exclusivity.
- Evidence-Based Governance: Top global experts provide objective data for decision-making.
- Rivalry Between Major Powers: China and the U.S. are competing to establish their own governance frameworks.
- Increasing Difficulty in Reaching Consensus: A bipolar landscape makes it unlikely for global unified rules to be established.
For China, the success of WAICO depends on its ability to move from being a symbolic initiative to providing tangible benefits, such as building AI infrastructure and training talent for developing countries.
In Summary
Global AI governance is no longer about which country has the strongest technology; it's about which platforms are the most influential and which ecosystems are the most comprehensive. Both China and the U.S. are pursuing their own paths, with businesses and multilateral organizations also aligning with different approaches. The question for ordinary people is whether AI will replace jobs and whether they can access more inclusive AI services. These outcomes are closely tied to the governance framework in place.