On August 9, 2026, the AI-themed “AGI Bar” is located in Beijing’s technology-centered Zhongguancun district.
Pedro Pardo | AFP | Getty Images
Hello, I’m Evelyn. I am writing to you from Beijing. Welcome to the latest edition of The China Connection. This is a snapshot of what I’ve seen and heard from local businesses.
Is the AI race about who can spend the most (the US) or who has the lowest prices thanks to state-controlled electricity rates (China)? Maybe neither.
big story
For the Chinese government, the artificial intelligence race is a story of determination.
Bruce Liu, CEO of Esoterica Capital, said that if there is a dollar left in China, “that dollar will be spent on AI, not real estate.”
The goal is to become self-sufficient in AI without relying on the United States, he said. “They don’t need to have the best AI in the world.”
Everything from China’s national policies to district-level subsidies supports these ambitions. So far, China has made progress with advanced chips to power AI. However, it still falls short of Nvidia.
now Nvidia It is rallying Wall Street giants to help raise $500 billion for AI development, demonstrating America’s superiority in capital.
Private sector AI investment in the U.S. is about 23 times higher than in mainland China, according to Alexander Keder, TMT analyst at BMI, a Fitch Solutions company.
Unless Beijing makes it easier for Chinese AI companies to leverage foreign non-state capital, “this funding asymmetry will remain one of the most persistent structural explanations for U.S. leadership,” Kedar said.
That hasn’t stopped Chinese companies from releasing AI models with similar functionality at lower prices, despite DeepSeek’s weekend price increase. Companies all over the world want to try these out.
But the model still requires chips to run, and the Chinese government lacks that capability compared to the United States.
Clifford Kurz, director at S&P Global Ratings, said China “could announce further financial support.” “But if they don’t have the chips, there’s no point in helping them. There’s nothing to fund.”
Kurz said Huawei provides only about one-eighth of Nvidia’s computing power, most of it outside China. He pointed out that Huawei’s most advanced Ascend 950 chips each have about 13% of the computing power of a single Nvidia GB300 chip.
Nvidia is launching an even more powerful Vera Rubin chip this year. And for now, Huawei is making up for it by piling on more chips.
But Kurz noted that Chinese companies are expected to produce just 1.35 million advanced AI chips this year, far less than the most conservative estimate of 6 million Nvidia chips.
follow the return
The story could change quickly. In just a few years, Huawei and other Chinese companies along the AI supply chain have closed the gap with their global rivals. China is also looking for AI talent, and electricity costs are low.
For investors such as Raffles Family Office, China’s domestic semiconductor drive is creating a “parallel” opportunity rather than a race for capital for U.S. tech, said William Chou, group deputy CEO of the Hong Kong and Singapore-based firm.
He said customers are much more concerned about the entry price of their AI investments this year. Chou noted that Nvidia’s financing plan shifts much of the risk from equity to credit, which means investment diversification is more important.
This is where the conversation once again deviates from America.
Regarding China’s AI projects, “we have not observed any significant plans for large-scale bond issuance by major domestic companies,” said Zhu He, a senior researcher at the Beijing-based economic think tank CF40 Institute. According to CNBC’s Mandarin translation:
Zhu said most companies are using equity financing or internal funds for AI spending, and not only internet companies but also telecom giants are investing in the technology.
In June, China announced a three-year plan to build infrastructure to support faster computing capacity, and last month said it expected to raise 4 trillion yuan in capital by 2030 to build a network of domestic computing capacity.
In both China and the United States, the scale of funding needed for AI reflects a shift away from the asset-light model that has helped companies win over the past two decades, said Esoterica’s Liu. “Hyperscalers need to spend money to move forward.”
Ultimately, AI still faces commercialization tests. While US companies have spent millions developing the smartest models, China’s goal is AI integration across industries.
“The AI race is about applications based on the complete AI stack,” said Winston Marr, an adjunct professor at New York University School of Law.
The one who finds the correct formula is ready to win.
need to know
The Chinese government is said to be working to clarify the tax system, which is causing confusion among China’s ultra-wealthy.
China’s State Administration of Taxation is conducting training for local tax officials to fine-tune the technical details of how to apply tax to offshore trusts, according to tax advisors.
Manas returns as an independent company
“We will soon resume operations as an independent company,” Manas said after Chinese regulators in April demanded that Meta withdraw its $2 billion acquisition of the artificial intelligence startup.
China’s answer to Boeing and Airbus takes first international flight
The Chinese-made C919 has begun a commercial flight route between Beijing and Ulaanbaatar, Mongolia.
Tencent anticipates spending surge, defends potential ‘superior’ AI returns
The Chinese tech giant announced a 65% increase in capital spending in the June quarter as it continues to invest in AI infrastructure. Tencent added that domestic gaming revenue increased 17% year-on-year, accelerating from the first quarter.
very soon
Unitree IPO is scheduled for this week
August 18: Pony.ai earnings
August 19th to 23rd: World Robotics Congress in Beijing
August 20: People’s Bank of China monthly decision on loan prime rate indicators
August 20: Alibaba earnings
August 22nd to 26th: World Humanoid Robot Games in Beijing
