American technology giant Oracle has reportedly cancelled a $7 billion agreement to provide China’s Tencent with access to advanced artificial intelligence chips, following pressure from the Trump administration.

The decision highlights Washington’s growing efforts to restrict Chinese access to American AI technology, including computing resources hosted outside the United States.

According to a Nov. 4 Financial Times report, Oracle had agreed to lease Tencent access to approximately 100,000 advanced AI chips through data centers in Southeast Asia. The five-year contract would have been Tencent’s largest overseas computing agreement.

However, U.S. officials reportedly approached Oracle with concerns about the arrangement, while Chinese authorities separately pressured Tencent to prioritize domestically produced semiconductors.

Reuters reported that it could not independently verify the cancellation. Oracle, Tencent, and the White House did not immediately respond to requests for comment.

For the best of our weekly content!

There is something wrong, please try again later
By signing up, you agree to our Terms of Service and Privacy Policy, and to receive messages from Vision Times.
SIGN UP

You are now signed up for our newsletter

Check your email to complete sign up

Washington moves to restrict China’s access to overseas AI computing

The cancelled agreement would have provided Tencent with access to advanced processors that are difficult to obtain within China because of American export restrictions and Beijing’s efforts to promote domestic alternatives.

Chinese technology companies have increasingly turned to overseas data centers to obtain computing power without physically importing restricted chips.

These arrangements have become particularly important for companies developing large language models, which require enormous amounts of processing power for training and operation.

Tencent, which operates the popular WeChat platform, has been investing heavily in artificial intelligence as it competes with Chinese rivals Alibaba and ByteDance.

During an August earnings call, Tencent Chief Strategy Officer James Mitchell emphasized the company’s commitment to developing its own AI models.

“We’re allocating a very substantial proportion of the new compute to building our own models to state-of-the-art status,” Mitchell said, according to the company’s earnings transcript.

The Oracle agreement would have substantially expanded Tencent’s access to advanced computing infrastructure as it developed its Hunyuan AI models.

However, Washington has increasingly scrutinized arrangements that allow Chinese companies to benefit from American semiconductor technology through foreign facilities.

During an April 22 congressional hearing, U.S. Commerce Secretary Howard Lutnick reiterated the administration’s position on advanced chip exports.

“I want to be crystal clear: We are not selling our best chips to China under any circumstance,” Lutnick told lawmakers, according to the Washington Examiner.

The reported cancellation suggests that American officials are increasingly concerned not only about physical semiconductor shipments to China but also about Chinese companies accessing powerful computing systems through overseas cloud services.

Nvidia founder and CEO Jensen Huang introduces the Rubin GPU and the Vera CPU as he speaks during Nvidia Live at CES 2026 ahead of the annual Consumer Electronics Show in Las Vegas, Nevada, on Jan, 5, 2026. (Image: Patrick T. Fallon / AFP via Getty Images)

Nvidia tightens cloud restrictions as Beijing pushes domestic alternatives

The Oracle decision follows reports that Nvidia, the world’s leading AI chipmaker, has begun restricting how much business its cloud computing partners can conduct with Chinese customers.

According to a Nov. 3 Financial Times report, Nvidia has instructed cloud providers to ensure that at least 51 percent of their services are sold to non-Chinese companies.

The reported guidance applies to future agreements rather than existing contracts and was communicated verbally through Nvidia’s sales teams.

Although Nvidia did not explicitly confirm the restrictions, the company defended the strategic importance of American-built AI infrastructure.

“Datacenters and clouds built on American chips are the foundation that American model and application makers need to compete worldwide,” Nvidia said in a statement reported by AFP.

The company added: “America cannot win the AI race unless America’s AI infrastructure wins worldwide.”

The developments could create further obstacles for Chinese technology companies seeking access to advanced American processors.

Beijing, meanwhile, has been encouraging domestic technology companies to reduce their dependence on foreign semiconductor suppliers.

According to the Financial Times, officials in China’s Guangdong province visited Tencent and urged the company to support Chinese chipmakers rather than rely on overseas computing arrangements.

Chinese Foreign Ministry spokesperson Mao Ning responded to questions about Nvidia’s reported restrictions on Nov. 3, saying she was unaware of the specific developments.

“As a matter of principle, we consistently hope that all parties can provide a fair, transparent and non-discriminatory business environment for companies from all countries, including Chinese companies,” Mao said, according to AFP.

Despite the growing restrictions, Chinese technology companies remain major customers of Southeast Asian data centers. Alibaba and ByteDance continue to lease substantial computing resources from overseas providers, while Oracle maintains other business relationships with Chinese companies.

The collapse of Tencent’s reported $7 billion deal illustrates how the competition between Washington and Beijing is extending beyond semiconductor exports into the global cloud computing industry.

Original article