Their tariff war might be stalemated, but the competition for technological supremacy between the US and China is shifting into high gear. As the two countries battle for dominance in artificial intelligence (AI) — and the productivity and geopolitical gains that would come with it — one question looms large: Would China’s AI capabilities catch up with — and even surpass — those of the US?
Driving this trend is a series of policies introduced by US President Donald Trump’s administration. Trump’s presidency marks a dramatic break from the commitment to openness that has underpinned the US’ technological leadership for decades. Measures intended to bring innovation back to the US might boomerang and end up paving the way for Chinese dominance.
The evolution of the digital economy might provide some insight into how today’s AI race would play out in the wake of Trump’s policies. In the 1990s, the US led the Internet revolution, dominating the pivotal “zero to one” phase by quickly moving innovations from lab to market. This fueled what many at the time lauded as the “new economy,” characterized by rapid growth, strong productivity gains and low inflation. China, initially a follower, later injected remarkable dynamism into the digital economy by scaling its own innovative technologies.
China’s digital development unfolded in three stages. The first was copy-and-follow: From the mid-1990s to the early 2000s, Chinese firms mirrored US models, launching Web portals and online services that drove explosive user growth.
The second stage was localization and improvement. As China’s digital ecosystem matured between 2005 and 2015, Chinese tech companies began to leverage their deep understanding of domestic users and market conditions to fine-tune their services. Platforms such as WeChat and Taobao not only adapted US concepts but also built on them, eventually surpassing Western counterparts such as WhatsApp and eBay in the Chinese market.
The third stage has been marked by breakout innovations. Over the past decade, Chinese tech companies have shifted from imitation to innovation, pioneering new digital models and overtaking foreign competitors. The most strikingly successful example is ByteDance’s TikTok, which positioned China at the forefront of online culture, reshaped social media, and forced US firms such as Meta to play catch-up.
CHINA’s CAPABILITIES
This dynamic is already evident in fields such as renewable energy and electric vehicles (EVs), and AI would be no exception. Following the launch of ChatGPT in late 2022 — which arguably marked AI’s transition into its mass-adoption era — China quickly demonstrated its ability to copy Western models.
The release of DeepSeek in January signaled China’s entry into the localization and improvement stage, as the company’s R1 model was 30 to 50 times cheaper to use than that of OpenAI. By February, the performance gap between the best Chinese and US models had narrowed to 1.7 percent, down from 9.3 percent last year. While it took ChatGPT two months to reach 100 million active users, DeepSeek reached that milestone in just seven days.
One of China’s key advantages is its deep pool of engineering talent. The country produces four times as many science, technology, engineering and mathematics graduates annually as the US. Beyond sheer size, this “engineer dividend” reflects a strong work ethic and a pragmatic mindset geared toward complex, hands-on optimization, as demonstrated by DeepSeek’s system architecture.
With more than 1 billion Internet users and a diverse industrial base, China also offers unparalleled conditions for deploying, testing and refining AI applications. China accounts for about 30 percent of global manufacturing output, generating vast amounts of data. In 2019 alone, its manufacturing sector produced 1,812 petabytes of data, and we estimate that figure reached 2,435 petabytes last year.
TRUMP’S CUTS
Energy is another critical factor. In 2023, China generated approximately 9,456 terawatt-hours (TWh) of electricity — 32 percent of the global total and more than double the US output of 4,178 TWh — giving it a major advantage in powering the large-scale data centers essential to widespread AI adoption.
The US’ position in the AI race is further undermined by Trump’s cuts to research funding and immigration restrictions. For example, in February the Trump administration laid off 170 employees, including AI experts, at the US National Science Foundation and proposed cutting the agency’s budget by more than 50 percent.
These cuts — together with the US National Institutes of Health’s delayed funding allocations and the freezing of about US$2.2 billion in federal grants to Harvard University — risk stalling foundational research and impeding AI innovation. Meanwhile, restrictive immigration policies would likely make it harder for the US to attract and retain global talent, potentially triggering a reverse brain drain as skilled Chinese tech workers return home to take up well-paid positions in a growing sector.
While the Trump administration has backed massive infrastructure initiatives such as Stargate — a proposed US$500 billion AI data center to be built by OpenAI, Oracle and SoftBank — such projects risk reinforcing big tech’s dominance and stifling the innovation needed to achieve transformative technological breakthroughs.
However, the deeper issue lies in the US’ shift away from economic openness. As US companies such as OpenAI become increasingly closed, Chinese firms are embracing open-source strategies. While Trump’s trade and immigration policies are driving away global talent and international collaborators, China is marketing its low-cost AI models to its trading partners.
China doubtless faces its own internal challenges, compounded by US trade restrictions that have limited its access to advanced semiconductors. Domestically, Chinese policymakers must strike a delicate balance between encouraging innovation and enforcing strict data controls. However, while neither side has an easy path to AI dominance, Trump’s “make America great again” agenda might inadvertently help make China great again.
Xu Qiyuan, senior fellow and deputy director of the Institute of World Economics and Politics at the Chinese Academy of Social Sciences, is the author of many books, including Reshaping the Global Industrial Chain: China’s Choices. Wang Yaqiang is an academic at the National University of Singapore’s Lee Kuan Yew School of Public Policy.
Copyright: Project Syndicate
A response to my article (“Invite ‘will-bes,’ not has-beens,” Aug. 12, page 8) mischaracterizes my arguments, as well as a speech by former British prime minister Boris Johnson at the Ketagalan Forum in Taipei early last month. Tseng Yueh-ying (曾月英) in the response (“A misreading of Johnson’s speech,” Aug. 24, page 8) does not dispute that Johnson referred repeatedly to Taiwan as “a segment of the Chinese population,” but asserts that the phrase challenged Beijing by questioning whether parts of “the Chinese population” could be “differently Chinese.” This is essentially a confirmation of Beijing’s “one country, two systems” formulation, which says that
“History does not repeat itself, but it rhymes” (attributed to Mark Twain). The USSR was the international bully during the Cold War as it sought to make the world safe for Soviet-style Communism. China is now the global bully as it applies economic power and invests in Mao’s (毛澤東) magic weapons (the People’s Liberation Army [PLA], the United Front Work Department, and the Chinese Communist Party [CCP]) to achieve world domination. Freedom-loving countries must respond to the People’s Republic of China (PRC), especially in the Indo-Pacific (IP), as resolutely as they did against the USSR. In 1954, the US and its allies
Indian Prime Minister Narendra Modi arrived in China yesterday, where he is to attend a summit of the Shanghai Cooperation Organization (SCO) with Chinese President Xi Jinping (習近平) and Russian President Vladimir Putin today. As this coincides with the 50 percent US tariff levied on Indian products, some Western news media have suggested that Modi is moving away from the US, and into the arms of China and Russia. Taiwan-Asia Exchange Foundation fellow Sana Hashmi in a Taipei Times article published yesterday titled “Myths around Modi’s China visit” said that those analyses have misrepresented India’s strategic calculations, and attempted to view
When Chinese President Xi Jinping (習近平) stood in front of the Potala Palace in Lhasa on Thursday last week, flanked by Chinese flags, synchronized schoolchildren and armed Chinese People’s Liberation Army (PLA) troops, he was not just celebrating the 60th anniversary of the establishment of the “Tibet Autonomous Region,” he was making a calculated declaration: Tibet is China. It always has been. Case closed. Except it has not. The case remains wide open — not just in the hearts of Tibetans, but in history records. For decades, Beijing has insisted that Tibet has “always been part of China.” It is a phrase
RSS