The United States may change its battle with China to dominate artificial intelligence

At the end of this week, Shanghai hosted the annual “Global Artificial Intelligence Conference” in China, an Amnesty International Organization Gallery, packed with technology giants from China and the United States including Huawei, Tesla and Amazon.

The topic was “global solidarity in the era of artificial intelligence”, and Chinese Prime Minister Lee Qiang opened the conference with a sweetener an offerEstablishing a global cooperation organization for Amnesty International, which is likely to be its headquarters in Shanghai. The Chinese Ministry of Foreign Affairs has since issued a work plan calling for international cooperation in artificial intelligence through open societies, source and joint research.

While the messages of Amnesty International in China are starting like “artificial intelligence for all”, the United States is still divided into its battle tactic. The Trump administration has given a commercial approach to isolation worldwide, especially with China when it comes to prosecution and technology. But with the changes in recent policy, this hardline lower appeared as Washington is divided between two camps on how to deal with the battle for artificial intelligence dominating China: that is, whether it will continue with the approach of intense protection or joining China’s calls for solidarity.

A year tense in geosus artificial intelligence

Beijing’s invitation to the gathering fell behind a Chinese vision to cooperate from artificial intelligence in the middle of a year nervous in geosus artificial intelligence.

The global leading United States was to develop artificial intelligence, but local confidence in America’s competitive advantage was shaken earlier this year. In the aftermath of Deepseek from Deepseek, the Trump administration has made a difficult line on advanced technology exports to China.

The administration tried to close the brakes to access devices in Beijing by increasing the current export controls on advanced NVIDIA chips to China, in an attempt to reduce the country’s rapid innovation, starvation of its ecosystem of artificial intelligence, and to maintain the hegemony of the United States.

But the ban did not go exactly to plan.

A Financial Times report Last week, it revealed that the advanced B200 chips that were banned in NVIDIA have almost smuggled to China in the three months since the export controls entered.

The administration changed the path around the ban and quietly retreated from its solid position earlier this month, when the CEO of Nvidia Jensen Huang announced that the company would resume selling the oldest H20 chips to China.

It is worth noting that FT has found that the demand for the Chinese black market on the smuggled B200s faced a noticeable decrease after relaxing in the H20 ban, indicating that Chinese companies prefer legal access to old chips on the highest technology illegally (companies that buy lumpy chips cannot benefit from the support of important NVIDIA customers when installed in their data centers). The effects of this discovery can throw the logic of a comprehensive export ban subject to question.

Hegemony through solidarity

China does not propose global solidarity in developing artificial intelligence from goodness in its heart. Open cooperation and joint research are not only the development of artificial intelligence technology in China, but also a soft power game by China. By expanding this cooperation in Shanghai and under Chinese terms and values, Beijing is trying to strengthen its position in the global artificial intelligence trade, and perhaps achieving AI global dominance over the United States

But Trump made it clear that he wanted America to win that battle.

“America is the country that started the artificial intelligence race and head of the United States, I am here to announce that America will win it,” Trump said during the announcement of measures last week.

The US government’s fears are two parts when it comes to AI and China: the loss of the economic edge and endangering national security.

Currently, the Chinese artificial intelligence industry depends on American chips such as NVIDIA. Smred in the banning of abdominal export says that if China has no access to advanced artificial intelligence technology, it will have no choice but to develop it. If China is building a real competitor to NVIDIA and acquires self -sufficiency in artificial intelligence devices, the United States may lose its grip on the global artificial intelligence market. At the center of these concerns, there is a Chinese technology giant Huawei, which is already developing intelligence computing systems The most advanced NVIDIA products in NVIDIA.

The advocates of this approach in Washington hope to have more control over the scale of innovation in China by flooding the market with American products. Especially by controlling what the chips go to the country, the United States can help reduce the spread of the most advanced chips technology in China. The last step of the Trump administration can be followed to reduce the restrictions imposed on the old NVIDIA chips models this logic, and may carry its fruits, at least, according to the results of the Financial Times.

The issue of national technical

They are two steps forward and one step.

Although the United States seemed to rest the rules related to the export of chips in China, the Trump administration is keen to continue the government’s solid line based on the president Artificial Intelligence Action Plan This was revealed last week.

“The advanced AI account is necessary for the era of artificial intelligence, which allows both economic dynamics and new military capabilities,” the administration wrote in the action plan. “The deprivation of our foreign opponents of reaching this resource, then, is the issue of competition of gesterodes and national security. Therefore, we must follow the creative methods to enforce export monitoring.”

The most stringent export control supporters have major national security concerns when it comes to developing artificial intelligence in China.

The Chinese AI companies have proven that you do not need the latest devices to make artificial intelligence that exceeds the standards, with the release of both the Kimi K2 supported by Deepseek and Alibaba this year. Although the latest chips are kept outside China while the elderly are dominated by the sales of the chips, this does not necessarily mean that the models of Amnesty International cannot be developed or even exceeding American models. Even after the market competition only, these models can be security risks when spreading for use in the Chinese army.

While China and the United States are not in a direct military conflict, the tensions are high between the great powers, especially on China’s regional demands in Taiwan and the South China Sea and America’s involvement in the region.

Commercial conversations and the following stage

All this comes at a time when the commercial envoys in the United States and China in Stockholm this week meet to discuss what is the exact long game here. The two countries are expected to decide the customs tariff agreement or choose another extension of the previously granted Indians that are scheduled to end on August 12.

As a result of these conversations, regardless of what, it is likely that they have effects that exceed what is mentioned on paper. It can be very good in the next stage of the war on global artificial intelligence dominance.