China's AI Ambitions Hit the Chip Wall

Beijing has made its bet. If China has one dollar left, "that dollar is going to be spent on AI rather than real estate," said Bruce Liu, CEO of Esoterica Capital. The goal is self-sufficiency, not necessarily global dominance. "They don't need to have the best AI in the world," Liu said.

Private sector AI investment in the U.S. runs about 23 times higher than in mainland China, according to Alexander Kheder, TMT analyst at BMI, a unit of Fitch Solutions. Unless Beijing makes it easier for Chinese AI firms to tap external non-state capital, "this financing asymmetry will remain one of the most durable structural explanations for US leadership," Kheder said. Nvidia has gathered Wall Street titans to support $500 billion in financing for AI development.

China has released cheaper AI models, including DeepSeek. Businesses globally are trying them. Running those models still requires chips. That is where Beijing falls short.

Huawei offers roughly one-eighth the computing capacity Nvidia has, mostly outside China, said Clifford Kurz, director at S&P Global Ratings. Each of Huawei's most advanced Ascend 950 chips has around 13% the computing power of one Nvidia GB300 chip. Nvidia has an even more powerful Vera Rubin chip coming this year. Huawei has compensated by piling more chips together. Kurz expects the Chinese company to produce just 1.35 million advanced AI chips this year – far less than even the most conservative estimate of 6 million Nvidia chips.

"China could announce even more financial support," Kurz said. "But if they don't have the chips, what's the point of support? There's nothing to finance."

The story could change quickly. Huawei and other Chinese companies along the AI supply chain have narrowed the gap with global rivals in a few years. China is courting AI talent and has low electricity costs.

For investors such as Raffles Family Office, China's domestic semiconductor push creates a "parallel" opportunity rather than competition for capital headed for U.S. tech, said William Chow, deputy group CEO of the Hong Kong- and Singapore-based firm. Clients care far more this year about the entry price for investing in AI, he said. Nvidia's financing plan shifts more of the risk to credit from equity, which makes diversification more important.

On debt financing for AI, the U.S. and China diverge sharply. "We have not observed any significant plans for large-scale debt issuance by leading domestic companies," said Zhu He, senior fellow at the CF40 Institute, a Beijing-based economic think tank, according to a CNBC translation of Mandarin. Zhu said most Chinese companies use equity financing and internal funds for AI spending. Telecommunications giants and internet companies are investing.

China in June released a three-year plan for building computing power infrastructure. Last month it said the buildout of computing power networks would attract 4 trillion yuan in capital through 2030.

The scale of money needed for AI reflects a shift away from the asset-light models that helped businesses win over the past two decades, said Esoterica's Liu. "Hyperscalers need to spend to get ahead."

Whichever country finds the right formula wins. "The AI rivalry is about applications based on the full AI stack," said Winston Ma, adjunct professor of law at New York University.

Separately, Tencent reported capital expenditures for the June quarter rose 65% as it continues investing in AI infrastructure. Domestic game revenue jumped 17% year-on-year, accelerating from the first quarter. The company has an Alpha Score of 42, labelled Mixed.

Key dates: the People's Bank of China makes its monthly decision on the benchmark loan prime rate Aug. 20. The World Robot Conference runs Aug. 19-23 in Beijing.