China has launched the World Artificial Intelligence Cooperation Organization, or WAICO, as part of its push to shape global AI rules. The new body could deepen the divide between Chinese and Western approaches to AI governance.
What is WAICO? China’s new global AI body, member countries and impact explained
On 17 July 2026, Chinese President Xi Jinping opened the World AI Conference in Shanghai by announcing the creation of the World Artificial Intelligence Cooperation Organization (WAICO), calling it “an important milestone in the history of AI development.” A day earlier, representatives of 29 countries had signed the founding agreement in the same city, with UN Secretary-General António Guterres looking on. The new body will be permanently based in Shanghai.
It is tempting to treat the announcement as ceremony. That would be a misreading. WAICO’s immediate powers are modest, but it marks China’s most serious attempt so far to move from following technology rules written by others to writing those rules itself. This article examines what was announced, why China built the organization, and what its consequences are likely to be.
What Was Actually Announced
WAICO is an intergovernmental body, meaning an organization whose members are national governments. Its founding agreement commits it to the purposes of the UN Charter and to putting people, rather than any single state’s interests, at the center of AI development. The idea is not new. Premier Li Qiang first proposed such a body at the 2025 edition of the same conference, and Xi repeated the proposal at the APEC summit in October 2025. What changed last week is that a proposal became a signed institution.
The founding membership tells its own story. It includes Russia, Belarus, Serbia, Cuba, Brazil, Venezuela, Pakistan, Kazakhstan, Indonesia and Laos, along with ten African and twelve Asian countries. No major Western economy, not the United States, the United Kingdom, Germany, France, Japan or Australia, signed on. Xi paired the launch with concrete offers: 5,000 AI training and seminar places for developing countries over the next five years, and deeper cooperation with regional groups such as ASEAN, the African Union, the Arab League and BRICS.
Why China Wants to Lead
Three factors explain the move. The first is capability. By Beijing’s own figures, China’s core AI industries were worth over 1.2 trillion yuan (about 176.6 billion US dollars) in 2025, the country hosts more than 6,200 AI companies, and its open-source models, systems whose code and weights are free to download and adapt, have been downloaded more than 10 billion times. China is not asking for a seat at the table; rather, it is working to become the head of the table.
The second is an opening. The United States has leaned toward light regulation at home and strict export controls abroad, arguing at a recent UN dialogue that heavy rules would stifle breakthroughs. The European Union passed the world’s first comprehensive AI law in 2024, but its approach has not been widely copied in Asia. In short, no agreed global rulebook or architecture exists, and through WAICO, China wants to gain first mover advantage and hence leadership of AI governance.
The third is the audience. Existing clubs such as the G7’s Hiroshima Process and the OECD’s AI principles tie membership to shared democratic values; a country joins because of what it already believes. WAICO’s offer is different: open to any state, and focused on development, training and access rather than values tests. For developing countries that feel priced out of the AI race, that is an attractive pitch. Xi sharpened the contrast himself, saying AI development should not be one country’s solo act but “a symphony of international cooperation”, and warning against stretching national-security arguments in AI, remarks widely read as aimed at Washington, though he named no country.
The Likely Ramifications
Four consequences appear most likely. First, a split rulebook. Countries in WAICO’s orbit will have little need to align their AI rules with the EU AI Act, OECD principles or G7 codes, which points toward two parallel governance systems. For companies operating across both blocs, that means duplicated compliance work; for smaller states, growing pressure to choose a side. The rivalry already has two named poles: Washington’s “Pax Silica” initiative, focused on securing chips and infrastructure for allies, and Beijing’s WAICO, focused on norms and standards, one seeks to control the pipes, the other to write the rules.
Second, technology adoption is likely to follow training. Thousands of training places, an AI weather-observation system offered to 30 countries, and free, capable Chinese open-source models together lower the cost for developing countries of building on Chinese technology. Once engineers are trained on particular tools and national systems run on particular models, switching becomes expensive. Influence over standards tends to follow such dependence, much as it did with earlier Chinese infrastructure lending.
Third, the organization’s limits are real. Writing in Nature, Elizabeth Gibney argued that WAICO is unlikely to govern the AI industry in any enforceable way, and the founding roster, Brazil and Indonesia aside, contains few large economies. Whether middle powers that trade heavily with both blocs eventually join, hedge, or stay out will be an early test of WAICO’s pull. A body without the US and EU is freer to act, but also poorer in leverage.
Fourth, the UN now sits in an awkward spot. Guterres’s attendance lent the launch visibility, and WAICO formally anchors itself in the UN Charter. Whether it feeds energy into UN-centered AI governance or draws attention away from it remains an open question.
Conclusion
WAICO’s importance lies less in what it can enforce today than in whom it convenes and what norms it seeds. Its future depends on delivery: if membership brings real computing power, models and training, influence over standards will follow; if not, it risks remaining a diplomatic showcase. Either outcome points the same way. The era in which the world might have agreed on a single set of AI rules appears, for the foreseeable future, to be over.
