Alibaba AI cloud revenue rose 53% as capital spending surged

Chinese technology firms are bracing for impacts on their ambitious artificial intelligence plans as the United States tightens restrictions on China’s access to the world’s cutting-edge chips. The new rules could affect some of Alibaba’s global data-center expansion plans and leave Tencent with less computing capacity for training newer models.

Alibaba, which on Thursday announced another quarter of strong revenue growth for its global AI cloud and compute services, reported that revenue from those services rose 53% from a year earlier to nearly 61 billion yuan ($9 billion) for the July-September quarter. Total revenue for the quarter rose 10% from a year earlier to almost 272 billion yuan ($41 billion), slightly better than analysts expected.

Alibaba’s capital expenditures rose 60% during the quarter to 50.5 billion yuan ($7.6 billion) as the company increased spending on AI infrastructure buildout. The company also reported a strong cash buffer.

The technology company warned that the U.S. rules could affect some of its global data-center expansion plans and said it was still assessing the repercussions. As global AI usage grows, Alibaba has been expanding its capabilities across chips, cloud infrastructure and AI models. It launched its flagship Qwen4-Max AI model earlier in the week.

In September, Alibaba’s CEO called the company’s new Zhenwu V900 chip the “most powerful” AI chip in China, while Alibaba announced massive data-center investment plans in China and abroad. Alibaba Cloud’s AI-related revenue currently comes significantly from China. Jacob Cooke of the Beijing-based consultancy WPIC Marketing + Technologies said Chinese enterprises still represented a lot of “untapped demand.”

Tencent said on Wednesday the new U.S. rule on chip access “sets everyone back in China.” President Martin Lau said on an earnings call: “Everyone is actually having less compute, advanced compute, to work with, no matter when you actually started leasing or building capacities outside of China.” He said training timelines for the next versions of Tencent’s AI models would probably be affected.

The restrictions target a route some Chinese companies have used to bypass existing U.S. export controls. As the technology race between the two superpowers intensifies, those controls already barred China from obtaining the world’s most powerful chips. Some Chinese companies rented remote computing access to train AI models, pushing the U.S. to close what some policymakers called a loophole.

James Chai, a researcher at the ISEAS – Yusof Ishak Institute in Singapore, said “Chinese AI firms are still heavily dependent on the most frontier Nvidia chips.” He said the new rules on remote access were cutting off a huge portion of Chinese firms’ access to compute. A study released in October by the Institute for Progress, a U.S. think tank, estimated that remote access to AI chips through overseas data centers could account for roughly one-third of China’s AI compute by the end of 2027.

Nvidia said it was “no longer shipping” chips including the H200, one of its more powerful chips but not the most advanced, even though the Trump administration had greenlighted some sales to China. The change reflected Nvidia’s tougher stance on China following Jensen Huang’s departure.

The Trump administration issued the rules this week in the wake of Nvidia CEO Jensen Huang’s resignation. Reporting described an extramarital affair with Alice Huang, a former senior executive of Singapore-based cloud computing company Megaspeed, who was alleged to be a Chinese spy.

The New York Times earlier reported that U.S. officials suspected Alice Huang was a Chinese spy recruited in part to potentially influence Nvidia to make its most advanced chips more available to Chinese companies. Beijing rejected claims that Alice Huang was a Chinese spy and called the new U.S. rules blocking Chinese companies’ remote access to computing power overseas “economic coercion.” Analysts expected the United States to step up its chip export control policies against China after Huang’s relationship with Alice Huang raised U.S. national security concerns.

Analysts and industry leaders in China said tighter controls would ultimately accelerate the country’s push for a self-sufficient chip industry. Lau said: “Over the longer run, we believe all Chinese players would actually have to rely on domestic supply chains and domestic chips, which we actually see the industry is actually progressing very fast.”

Alibaba had not finished assessing how extensively the rules would affect some of its global data-center expansion plans. Tencent, meanwhile, said its next model-training timelines would probably be affected, while Lau expected Chinese AI companies to rely more heavily on domestic chips over the longer run.