Labs

Xi launches 29-nation
AI bloc from
Shanghai, positions China
as governance rival
to the West

In his first-ever WAIC keynote, Xi Jinping announced the World Artificial Intelligence Cooperation Organization — a Shanghai-headquartered intergovernmental body pitched to the Global South as an alternative to U.S.-led AI order.

By Casey Vanderhoof Frontier labs correspondent · July 17, 2026

Xi Jinping used his first-ever keynote at the World AI Conference on July 17 to launch the World Artificial Intelligence Cooperation Organization, a 29-nation intergovernmental body headquartered in Shanghai and pitched, in all but name, as the multilateral answer to a U.S.-led AI order that most of the Global South has watched from the outside.

The founding members had signed on a day earlier, per Xinhua. Indonesia, Brazil, Malaysia, South Africa, Senegal, Russia and Pakistan were among them, per Al Jazeera. Premier Li Qiang first floated the organization in July 2025, giving Beijing a full year to line up the roster before Xi walked on stage. UN Secretary-General António Guterres sat in the audience, a detail Beijing wanted photographed.

WAICO’s opening pitch is distribution. Xi pledged 5,000 AI training and seminar opportunities for developing countries, plus cooperation frameworks with ASEAN, the African Union, the Arab League and BRICS. This is the Belt and Road playbook rebuilt for the compute era: infrastructure, curriculum, standards, and a Shanghai postmark on all of it.

The subtext was hardware. On the WAIC show floor, Huawei’s Atlas 950 SuperPoD, a supernode designed to link 8,192 Ascend chips with no U.S.-origin components, sat as physical proof that export controls have a shelf life. NPR framed the keynote against Washington’s chip and tooling restrictions, arguing the curbs accelerated Beijing’s pivot from importer to system-builder. Bloomberg noted that Chinese models are approaching a record 60% share of U.S. firms’ AI usage on the OpenRouter marketplace. American developers are already voting with their API keys.

Xi’s rhetorical maneuver was the part worth rereading. He called for AI to remain “secure and controllable” and “always” under human control, language Bloomberg’s live coverage flagged as an unusually explicit endorsement of human oversight from Beijing. In the same breath he urged countries to oppose “overstretching the national security concept in the field of AI,” a line CNBC and Al Jazeera read as a jab at Washington. The choreography lets China claim the safety-pilled center of gravity while framing U.S. export controls as the actual destabilizing force.

That’s the governance play. For most of the last decade the assumption inside the Western policy apparatus was that AI norms would be written in Brussels, litigated in Washington, and grudgingly accepted in Beijing. WAICO inverts the sequence. It gives 29 states a venue where the drafting happens in Shanghai and the training seats are already funded.

Whether the organization becomes a real standards body or a diplomatic prop depends on what the second cohort of signatories looks like, and on whether Huawei’s supernode economics hold once the marketing slides come down. But the framing shift has already happened. Beijing is no longer arguing about access to the American stack. It’s building its own and inviting the rest of the room in.

Sources