China Pushes for a UN Role in AI Rules With Xiong'an Vision
China is pitching open source artificial intelligence and a louder voice for the Global South as the basis for future tech rules, in a move aimed at challenging US dominance in setting standards for the sector. The position was set out in a document called Global Governance, Xiong'an Vision, released at the close of a three day forum in Xiong'an New Area. The site is a planned city west of Beijing built under a directive from President Xi Jinping.
Delegates from more than 90 countries and international bodies attended, according to Chinese officials. The Global South here means developing economies outside the traditional Western sphere, spanning Latin America, Africa and parts of Asia and Oceania.
The document calls for international collaboration and responsible development of open source AI. And it asks for research findings and technical expertise to be shared, so that access to frontier technology widens beyond the countries that already hold it.
"It is important to take seriously and proactively address the technological and ethical risks posed by AI, preventing its misuse and making sure that it remains under human control," the text reads.
And the governance question sits at the centre of it. The document argues for the United Nations to play a central role in AI governance, on the grounds that multilateral oversight is the only way to give every nation an equal voice. That framing puts the UN, rather than any single government or industry body, at the head of the table.
But Klaus Schwab, founder of the World Economic Forum, backed the argument. He warned that rapid advances in AI and other emerging technologies have created an urgent need for new international frameworks. "Without China's participation, no global governance system can truly take shape," he was quoted as saying in a state media report.
But the backdrop is a contest, not a consensus. Beijing and Washington are locked in a race that goes beyond training the most capable models, and extends to the standards that decide how those models are deployed across borders. Whoever writes the rules shapes the market that follows.
The two sides start from different positions. US tech giants hold an edge in proprietary, closed source models, while Chinese contenders such as DeepSeek and Alibaba Group Holding have leaned heavily into open weight architectures. Alibaba owns the South China Morning Post, which reported the document.
And open weights let developers and enterprises download, adapt and build on a model without negotiating a licence. That has helped Chinese companies widen international adoption, and it has given Beijing a route to influence the global AI agenda that does not depend on having the strongest model.
The Xiong'an document leans on that logic. By framing open source as a public good and the UN as the rightful forum, China is appealing to countries that have no seat in the US led technology order and would gain from cheaper access to capable systems. Countries that cannot license the strongest closed models have reason to listen.
China released the Xiong'an Vision document at the close of a three day forum west of Beijing
It backs open source AI, shared research and a central role for the United Nations
Delegates from more than 90 countries and international bodies attended, Chinese officials said
Klaus Schwab said no global governance system can take shape without China
The push comes as Beijing and Washington compete over the standards that govern cross border AI use
Source: SCMP
