WAICO: 29 Countries Launch Rival AI Governance Body Without US

What happened

Twenty-nine countries — led by China and including Russia and Brazil — formally established the World Artificial Intelligence Cooperation Organization (WAICO) at a founding ceremony in Shanghai on 16 July 2025. The body is designed to set universal standards and introduce regulatory frameworks for artificial intelligence at an intergovernmental level.

The United States, the European Union and its member states, the United Kingdom, Japan and South Korea were all absent from the founding membership. Their exclusion is significant: WAICO represents a deliberate attempt by China to position itself as an alternative centre of gravity in global AI governance, at a moment when Chinese enterprises are actively working to close the gap with US competitors across open-weight models, open-source AI and AI-driven cybersecurity.

Researchers note that WAICO is structurally distinct from previous multilateral AI initiatives in at least three respects: membership is open to any sovereign state; there is no regime-based eligibility filter; and the organisation has been years in development rather than assembled reactively. Whether those design choices will attract broader participation — or deepen the fracture between competing AI blocs — remains to be seen.

Why it matters

For customer-experience practitioners and service designers, the emergence of a parallel AI standards body is not an abstract geopolitical story. The tools that power personalisation engines, conversational interfaces, fraud detection and predictive service models are increasingly shaped by the regulatory environments in which their developers operate. A world in which AI governance splinters into competing blocs means that the models underpinning customer journeys may be built to different safety, transparency and data-handling standards depending on their country of origin — creating real complexity for any brand operating across markets.

From a behavioural-economics perspective, the fragmentation also affects trust. Customers do not read standards documents, but they absorb the downstream effects: inconsistent explanations from AI agents, opaque decision-making in credit or service eligibility, and varying levels of recourse when automated systems err. Organisations that source AI capabilities globally will need to treat governance provenance as a service-design variable, not merely a procurement footnote.

The Renascence take

Most commentary on WAICO will focus on the geopolitics. The more consequential question for operators is quieter and more practical: when the AI powering your customer interactions was trained and governed under a different regulatory philosophy, whose definition of "fair," "transparent" or "safe" is your customer actually experiencing?

The instinct will be to treat AI governance as a compliance matter and hand it to legal. That is the wrong reflex. Governance shapes model behaviour, and model behaviour shapes the micro-moments that determine whether a customer feels understood or manipulated. Customer-obsessed operators should be auditing the provenance of every AI layer in their service stack now — not because regulators demand it, but because trust is built or destroyed at the interaction level, long before a standards body publishes its first framework. The fracturing of global AI governance is, at its core, a customer-experience risk that no CX team can afford to outsource entirely to procurement.