The first world body for AI governance is born in Shanghai — without the United States or Europe

On 16 July, on the sidelines of the World AI Conference in Shanghai, twenty-nine countries signed the founding act of the World Artificial Intelligence Cooperation Organization (WAICO), a new intergovernmental body headquartered in the Chinese city. Foreign Minister Wang Yi signed on behalf of Beijing; UN Secretary-General António Guterres also attended the ceremony. The stated aim is to "strengthen international cooperation and global governance on AI" so that it is "beneficial, safe and fair."

More telling than the words is the geography of the signatories. The founders include Russia, Pakistan, Indonesia, Kazakhstan and Laos, along with about ten African countries and several Asian states. Absent, however, are the United States and the European Union: the two powers that today hold the most advanced models and the strictest regulatory frameworks — from the European AI Act to American executive orders — remain outside the door. It is hard to read WAICO as anything other than an attempt by Beijing to build an alternative governance bloc anchored in the Global South.

In practical terms, though, many boxes remain unchecked. Details on the voting structure, funding and binding powers of the organization are not yet known: for now WAICO is a founding act and a headquarters, not yet an operating mechanism. And a body that presents itself as a global arbiter of AI without the two players that produce most of it starts from a weak negotiating position.

The signal remains, and it is not a small one: AI governance is fragmenting into clubs that eye one another with suspicion, each with its own idea of what "safe and fair" means. The risk is that, as the number of tables multiplies, common rules become ever harder to write. — Olya