China is running a different race.
In China, by contrast, leader Xi Jinping has recently had little to say about AGI. Instead, he is pushing the country’s tech industry to be “strongly oriented toward applications”—building practical, low-cost tools that boost China’s efficiency and which can be marketed easily.
The diverging visions represent a head-to-head bet with significant stakes. If China’s gamble turns out to be wrong, it could find itself lagging far behind the U.S. in the most consequential technology of the 21st century.
But if AGI remains a distant dream, as more people in Silicon Valley now believe, China will be in position to steal a march on its global rival in wringing the most out of AI in its current form, and spread its applications worldwide.
Already in China, domestic AI models similar to the one that powers ChatGPT are being used, with state approval, to grade high-school entrance exams, improve weather forecasts, dispatch police and advise farmers on crop rotation, state media and government reports say.
Tsinghua University, China’s equivalent of the Massachusetts Institute of Technology, is rolling out an AI-powered hospital, where human doctors will be assisted by virtual colleagues armed with the latest data on diseases. Intelligent robots are being deployed to run automotive “dark factories” and inspect textiles for flaws while still on the loom.
“They see highly impactful AI applications not as something to theorize about in the future but as something to take advantage of here and now,” said Julian Gewirtz, a former National Security Council official who specialized in tech competition with China during the Biden administration.
U.S. tech companies are developing plenty of practical applications using current AI, of course. Google has wired its latest Pixel smartphones to do real-time translation, while U.S. consulting companies are using AI agents to build PowerPoint decks and sum up interviews for clients. Others are using it to improve drug discovery and food delivery.
But unlike the U.S., which largely leaves the industry to its own devices, Beijing is putting the full muscle of the state behind its vision. In January, the central government unveiled an $8.4 billion AI investment fund focused on supporting startups. Local governments and state banks have since rolled out their own funding programs, while cities have published AI development plans as part of a campaign dubbed “AI+.”
On Tuesday, China’s cabinet spelled out broader ambitions for the campaign, calling for an even stronger push to integrate AI into science and tech research, industrial development and other areas to “comprehensively empower” China’s economic development by 2030.
China is also more actively embracing open-source models that are free for users to download and modify, making it cheaper and easier for Chinese companies to build businesses around the technology. That approach is helping Chinese AI spread globally, a trend that has shaken Silicon Valley into following suit.
AGI dreams
That emphasis is somewhat different from the ambitions of many of the U.S.’s biggest tech players, who believe that machines that can outthink humans will revolutionize science, open up entirely new fields of inquiry and transform the American military.
Some in the tech industry have predicted that artificial superintelligence could arrive as soon as 2027. Companies such as Meta, Google and OpenAI are spending lavishly in a competition to acquire the talent, data centers and energy they need to be first.
A congressional commission focused on competition with China has floated a “Manhattan Project” for AGI to ensure the U.S. wins the race.
But OpenAI’s highly anticipated release in August of GPT-5, a model the company had initially touted as a major steppingstone on the path to AGI, left many users underwhelmed. OpenAI founder Sam Altman acknowledged the bumpy rollout and has since tried to tap the brakes on AGI hype and warned about the possibility of an AI investment bubble.
Other Silicon Valley titans have also started to waver, opening the door to the idea that China’s approach might make more sense.
“It is uncertain how soon artificial general intelligence can be achieved,” former Google Chief Executive Eric Schmidt and technology analyst Selina Xu wrote in a recent opinion column for the New York Times.
“In being solely fixated on this objective, our nation risks falling behind China, which is far less concerned with creating A.I. powerful enough to surpass humans and much more focused on using the technology we have now.”
Pragmatic approach
The Chinese government’s enthusiasm for more-practical uses of AI is visible in Xiong’an, Xi’s built-from-scratch dream city two hours south of Beijing.
In February, the city announced the release of an agricultural AI model, using technology from Chinese startup DeepSeek, that gives local farmers guidance on crop selection, planting and pest control, according to a local government report. The city’s meteorological service is using DeepSeek to improve the accuracy of weather reports. DeepSeek is also helping local police analyze case reports and decide how to respond to emergencies.
Xiong’an’s branch of 12345, a government question hotline that fields hundreds of thousands of calls a day nationwide, is using DeepSeek to sort and route inquiries.
A major portion of government investment is going to build data centers. But unlike the sprawling facilities being built in the U.S. to train cutting-edge models, the Chinese versions tend to be smaller.
To a large extent, Beijing has no choice but to break a different trail on AI. U.S. trade restrictions, particularly on high-end semiconductors, have made it difficult for Chinese AI companies to compete head-to-head with American giants in scaling up the training of the most advanced models.
The choice makes even more sense given growing uncertainty about the return on investment of chasing scale, said Jeffrey Ding, a professor at George Washington University and author of ChinAI, a newsletter focused on Chinese AI.
“You let the technology leader, the U.S. in this case, eat the cost of exploration, and then you try to be the fast follower or be the one who optimizes for implementation,” he said.
To be sure, some Chinese companies, including DeepSeek and Alibaba, have said they would pursue AGI. And some analysts have speculated that China could be trying to keep a lid on some of its AGI ambitions.
Chinese tech company Alibaba has said it is pursuing artificial general intelligence.
It is possible, even likely, that Xi will decide at some point to more aggressively pursue AGI, said Kendra Schaefer, head of tech-policy research at Trivium China, a Beijing-based consulting firm. But he will do so cautiously with plenty of safeguards, she said, given the potential risk that thinking machines could pose to Communist Party stability.
“It is one of the most risk-averse governments on the planet,” she said.
Like the internet, which had to weather the dot-com crash and years of development before it could rewire the global economy, it could take decades to determine winners and losers in AI, according to George Washington University’s Ding.
The U.S. has important advantages over China in harnessing new technologies, he said, including a broad education system beyond elite universities that can spread technical knowledge widely.
If it is careful to maintain that edge, Ding said, the U.S. has a good chance at eventually beating China at its own race.
Write to Josh Chin at Josh.Chin@wsj.com and Raffaele Huang at raffaele.huang@wsj.com
