China is narrowing the gap with the US in technological advancements, even with existing limitations, according to AI experts

China's Push to Close the Tech Gap with the U.S.

By Laurie Chen

China is making strides to catch up with the United States in technology, fueled by a surge in innovation and a greater appetite for risk. However, the country's progress is hampered by the absence of advanced chip manufacturing equipment, according to top artificial intelligence experts in China.

This week, Chinese AI startups MiniMax and Zhipu AI made impressive entries on the Hong Kong Stock Exchange, signaling increased optimism in the industry. The Chinese government is accelerating the listing of AI and semiconductor companies, aiming to strengthen homegrown alternatives to cutting-edge American technologies.

Yao Shunyu, who previously held a senior research position at OpenAI and was recently appointed as Tencent's chief AI scientist, expressed confidence that a Chinese company could become a global AI leader within three to five years. Nonetheless, he emphasized that the lack of state-of-the-art chipmaking machinery remains the sector's biggest technical obstacle.

Speaking at an AI conference in Beijing, Yao noted, "We currently benefit from robust electricity supply and infrastructure, but our main challenges are in manufacturing capacity—especially lithography machines—and in developing a strong software ecosystem."

Last month, Reuters reported that China had developed a prototype of an extreme-ultraviolet lithography machine, which could potentially produce advanced semiconductor chips on par with those made in the West. However, sources indicated that this machine has yet to manufacture functional chips and may not do so until around 2030.

Investment Disparities Remain

At the same Beijing conference, Yao and other industry leaders acknowledged that the United States still holds a significant edge in computing power, largely due to substantial investments in infrastructure.

Lin Junyang, the technical lead for Alibaba's Qwen large language model, remarked, "The U.S. computing infrastructure is likely ten to a hundred times larger than ours. Major players like OpenAI are pouring resources into next-generation research."

Lin added that Chinese companies face financial constraints, with most of their computing resources dedicated to delivering existing services. He made these comments during a panel at the AGI-Next Frontier Summit, organized by the Beijing Key Laboratory of Foundational Models at Tsinghua University.

Despite these limitations, Lin pointed out that China's resourcefulness has led to creative solutions, such as algorithm-hardware co-design, which allows AI companies to operate large models on more affordable, smaller-scale hardware.

Tang Jie, founder of Zhipu AI—which recently raised HK$4.35 billion through its IPO—also praised the growing willingness among young Chinese AI entrepreneurs to take bold risks, a quality often associated with Silicon Valley.

Fostering a Culture of Innovation

Tang emphasized, "If we can further improve the environment to give these talented, risk-taking individuals more time and space to innovate, it would be a significant step forward—something both the government and the country can support."

Reporting by Laurie Chen; Edited by Emelia Sithole-Matarise