SStanding on stage in Hangzhou, eastern China’s tech hub, Alibaba’s usually media-shy CEO made an attention-grabbing announcement. “Today the world is witnessing the beginning of an AI-driven intelligent revolution,” Eddie Wu said at a developer conference in September. “Artificial general intelligence (AGI) will not only enhance human intelligence but also unlock human potential, paving the way for the advent of artificial superintelligence (ASI).”
ASI, Wu said, “could create a generation of ‘super scientists’ and ‘full-stack super engineers'” who would “tackle unsolved scientific and engineering problems at unimaginable speed”.
Wu also announced plans to invest 380 billion yuan (£40 billion) in AI infrastructure over the next three years, news that sent Alibaba shares to their highest in almost four years.
Wu’s foray into the existential, technological-frontier rhetoric typically deployed by Western tech CEOs like OpenAI’s Sam Altman and DeepMind’s Demis Hassabis caught the attention of observers. “Wu’s ASI speech represents a breakthrough,” technology writer Aphra Wang wrote in her China AI newsletter. concurrent. “Major Chinese companies are beginning to articulate their own grand visions that contain glimpses of what the future holds.”
AGI, a theoretical state of AI where a highly autonomous system is capable of performing the work of a human, has become a subject of concern for US tech companies such as OpenAI and DeepMind. Many see it as the next frontier of civilization, and are in competition with each other and China to get there. In May, Brad Smith, Microsoft’s chairman, told a US Senate committee on AI that “the race between the United States and China for international influence will likely be won by the fastest first mover”.
Many people in Washington have internalized these fears. The US-China Economic and Security Review Commission has recommended That Congress “establishes and funds a Manhattan Project-like program dedicated to developing and achieving an artificial general intelligence (AGI) capability”. The Manhattan Project was a World War II-era research operation to create nuclear weapons.
In China, many viewed Wu’s speech as expressing a bold, eccentric tech company vision, but not one that was representative of China’s overall AI industry.
“There are certainly research groups in China working toward AGI. But most AI companies are working toward better applications,” said Ya-Qin Zhang, dean of Tsinghua University’s Institute for AI Industry Research and former president of tech company Baidu.
The combination of limited computing power, a pragmatic approach to technology, and a keen awareness of AI’s current potential has led China’s national AI policy to move toward real-life applications rather than frontier research.
In August, the Chinese government published its much-anticipated “AI+ Strategy.” The policy document outlines how AI could impact China’s development goals, such as using AI to improve medical diagnostics and make supply chains more efficient. But there was no mention of AGI in it.
Julian Gewirtz, former senior director for China and Taiwan at the White House National Security Council, said, “The Chinese government is focused on realizing the benefits of AI here and now and in the near future through the diffusion and application of AI in the economy, society, defense and other areas.” “Despite its goal of ‘catching up and surpassing’ the United States, we should not assume that the Chinese Communist Party has bought into the idea that AGI is imminent.”
“If you’re just looking at what’s been officially published… there’s no clear acceptance of AGI,” said Selina Xu, a China technology analyst. Xu said Chinese leader Xi Jinping has a history of prioritizing the material economy over more intangible forces.
“It’s a very different story than the AGI race as a lot of people in D.C. see it,” Xu said.
One of the biggest factors guiding this strategy is the fact that US sanctions have prevented Chinese companies from obtaining the world’s most sophisticated semiconductors, which are essential for advanced AI research.
Washington has banned the sale of high-tech microchips to China in an effort to curb the country’s AI development. Nvidia, the world’s leading chipmaker, developed more basic semiconductors specifically for the Chinese market. In December, Washington approved Nvidia’s second-most advanced chips, the H200s, for sale in China. But Beijing has reportedly told customs agents that the chips cannot be imported into China, as the government seeks to end the country’s reliance on foreign technology.
China emphasizes that “necessity is the mother of invention” and points to the success of companies like DeepSeek as proof that US sanctions will only stifle innovation. DeepSeek’s founder, Liang Wenfeng, is one of the few Chinese tech leaders who, like Alibaba’s Wu, has openly expressed interest in AGI.
But until China is able to produce its own advanced semiconductors on a large scale, most tech companies that find it more profitable to use hardware will already have to focus on AI applications rather than AGI.
Another factor guiding US-China technology competition is the availability of datacenters and the energy to power them. In November, Nvidia CEO Jensen Huang said that China “will win the AI race” because of its energy subsidies for datacenters.
The subsidy was reportedly introduced after Chinese tech companies complained about high electricity bills due to the domestic semiconductors they are forced to use, which are less efficient than Nvidia’s. In a sign of how determined China is to break its reliance on imported technology, Reuters reported that any datacenter receiving state funding can only use domestic chips.
Such measures would reduce Nvidia’s competitive advantage in China and boost domestic chip producers like Huawei.
Since 2021, China has reportedly invested $100 billion in supporting AI datacenters.
But there are signs that the rally may have been overzealous. A recent report from the China Academy of Information and Communications Technology said the utilization rate of AI datacenters nationwide was 32%.
In a recent op-ed for China Economic Weekly, Rao Shaoyang, director of the China Telecom Research Institute, wrote that in some areas of China, the computing power industry was operating similarly to China’s troubled property sector: build first, find buyers later. He cautioned against “blindly building intelligent computing centers” and said local computing power demand should be considered before building new datacenters.
Despite a surplus in more general computing power, many experts believe China still does not have chips sophisticated enough to explore leading-edge research into AGI. But analysts say the mood could change quickly.
“The current status quo is highly untenable, and Xi Jinping has clearly declared an ambition to lead the world in AI,” Gewirtz said. “So the fact that China understands that goal one way in this snapshot moment in time doesn’t give me any comfort that in a year they’re going to understand it the same way.”
Additional research by Lillian Yang
