China’s Xi Jinping wants AI to be open to the world and outside American control

Chinese President Xi Jinping made his first in-person appearance at his country’s biggest artificial intelligence event, the annual World Artificial Intelligence Conference in Shanghai, on Friday. In his long-awaited first speech, Xi called for more open source AI and cross-border cooperation, taking a starkly opposite stance from his American counterpart: the price lover President Donald Trump.

“We often say in China: ‘A single string cannot make music, and a single tree does not make a forest,'” Xi told the crowd in Shanghai, according to China Daily. “The development of AI should not be the work of a single country, but a symphony of international cooperation. »

Xi has reason to say this. China and the United States are locked in a battle for dominance in the global AI industry. With pioneering American labs like OpenAI and Anthropic and the industry’s leading chip supplier, Nvidia, the United States is currently in the lead. But Chinese companies are quickly closing that gap, offering open models that can outperform some of the closed offerings from big U.S. tech companies.

The shocking improvements in the performance of some of these models have led the Trump administration to take a tougher stance on AI trade with China, much to the dismay of chip giant Nvidia, which once considered China one of its largest markets. Nvidia CEO Jensen Huang has advocated for the normalization of trade relations between the two superpowers, arguing that the wide prevalence of American technology abroad, notably in China, would mean that their AI industry would be entirely dependent on American technology, thereby guaranteeing American hegemony.

At Nvidia’s first GTC held in Washington, DC late last year, Huang warned the politicians in attendance that this dynamic could be reversed at any time. With China’s renewed interest in open source AI, Silicon Valley’s complete withdrawal from the Chinese market would leave the United States “ill-prepared” for when Chinese software “permeates the world.” Huang’s advocacy for US domination in China appears to be Xi’s blueprint for Chinese domination on a global scale. But for now, Xi’s proposed conception of Chinese domination is essentially that no one dominates anyone.

“We must seize this rare and historic opportunity to encourage open source, openness, collaboration and sharing,” Xi said. “As a responsible major country, China is always committed to providing international AI-related public goods. »

China is not only calling for global solidarity, but basing AI cooperation on Chinese conditions. At last year’s Global AI Conference, Chinese government officials proposed the creation of a global AI cooperation organization headquartered in Shanghai. At this year’s conference, Beijing formally announced the creation of a new coalition, the Global Artificial Intelligence Cooperation Organization, and said it had 29 member nationsincluding some heavyweights like Brazil and Russia.

This emphasis on open cooperation and joint research kills two birds with one stone: it promotes the development of AI in China and ensures that global AI development is guided by Chinese principles and under China’s wings, thereby positioning the country as the global leader in the industry.

By investing particularly in economies in the South, where the AI ​​landscape is only developing, China is seeking to establish early control over AI industries across the world. Also on Friday, Xi pledged to provide developing countries with 5,000 AI training program opportunities over the next five years and develop international cooperation centers for AI applications in the Global South.

“This is a major step by China to respond to the call from the South and unite the international community to promote vigorous development and governance of AI,” Xi said on Friday. “This will be an important milestone in the history of AI development. »

Although Xi praises “true multilateralism” with all countries, reports suggest Beijing is singing a different tune behind the scenes when it comes to the United States. According to Reuters, Chinese officials met with the country’s major laboratories to discuss limiting overseas access to open source and proprietary models. Chinese authorities also recently forced Meta to dismantle its acquisition of Chinese AI startup Manus and suspend all data sharing. Chinese AI developers are also growing in number move from Nvidia to local silicon suppliers under the encouragement of the Chinese government.

So this Chinese vision for global AI cooperation may not necessarily include the United States, even if the Trump administration reverses its trade hardline.

On Friday, Xi also made the case for AI safety and took a more pro-regulation stance, highlighting another way he wants Beijing to distinguish itself from Washington under Trump, while also attacking his administration’s national security policy decisions.

Last month, Trump banned Anthropic from sharing Mythos, its latest model, which has powerful hacking capabilities, with foreign nationals inside or outside the United States. He has since partially lifted this banallowing the company to start sharing the model again with select businesses and agencies. It is reasonable to assume that Chinese companies and agencies are not on this shortlist, given that when Washington talks about “national security concerns” in the context of AI, they are most often referring to China.

“We need to establish laws and regulations, technology monitoring, early warning and emergency response systems to strengthen the safety line, prevent abuse and malicious use, and ensure that AI is always under human control,” Xi said. “In the meantime, we must jointly oppose overextending the concept of national security in the field of AI and putting the security of one country before that of others.”