China begins to snap at America’s heels in a tightening tech race
LineShine uses locally designed Chinese chips and ranks high in traditional TOP500 benchmarks, but only ranks fourth in tests that better reflect AI-related computing tasks. This says more about China’s determination to demonstrate technological self-sufficiency than its leadership in the AI race.
Even so, symbolism is important. China has stopped sending systems to the TOP500 after years of increasing US export controls on chips and computing technologies. The decision to return now shows increased confidence. Addison Snell of Intersect360 Research told Reuters he was surprised not that China had built the top-tier system, but that Beijing wanted public recognition of it.
The shock that shook Washington
Nothing has crystallized America’s concerns more than the rise of DeepSeek. America believed that US dominance in advanced chips would maintain a significant lead in artificial intelligence, but DeepSeek challenged this belief. The Chinese chatbot delivered performance that many observers considered comparable to leading American models, using far fewer resources. DeepSeek forced Chinese developers to be more creative because they did not have access to the most advanced hardware. This means that export controls on China may have accelerated rather than slowed China’s self-confidence.
The effects extend beyond a single company. According to a recent report from the BBC, China’s open source culture has enabled developers to quickly benefit from each other’s advances. Selina Xu, who studies Chinese AI policy, told the BBC that Chinese models may not always match the best American systems, but are often much cheaper. This changes the economics of competition.
Data from Stanford University’s 2026 Artificial Intelligence Index shows how quickly the gap is narrowing. It confirms that the US still leads in frontier model performance, especially in judgment-intensive benchmarks and multi-modal systems. However, the gap has narrowed significantly in areas such as natural language understanding and code generation. The gap between the top AI bots in the US and China is narrowing in Arena scores, a metric that shows the relative performance of large language models. In May 2023, the US’s top model, OpenAI’s GPT-4, maintained its lead with more than 1,300 Arena points compared to China’s less than 1,000 points. By March 2026, the gap had fallen from 300 to 39 points. Anthropic’s Claude Opus 4.6 was leading China’s Dola-Seed 2.0 by just 2.7% points.
More striking is China’s dominance in scale measurements. It is generating a larger share of global AI publications and patents, demonstrating both the breadth and depth of its research activity. In industrial distribution, China is a leader in robotics integration, especially in manufacturing and logistics. The report also highlights the increase in domestic artificial intelligence initiatives supported by state funding and regional innovation clusters. But the data also reveals limitations. Chinese models tend to lag a bit in terms of reliability and compliance, in part due to tighter content controls that limit educational diversity. Access to cutting-edge chips remains uneven, forcing companies to innovate around hardware bottlenecks. This is where techniques such as distillation and optimization become critical for the Chinese. America is still ahead, but the gap is no longer comfortable.
Robot racing becomes the next battlefield
If artificial intelligence software remains an area of American strength, robotics is increasingly setting the pace for China. BBC analysis describes the competition as a battle between AI “brains” and AI “bodies”. America continues to dominate many of the most advanced AI models and chips. China has gained a tremendous advantage in producing robots and using them on a large scale. The numbers are striking. According to the BBC, China currently has around two million working robots; That’s more than the rest of the world combined. It accounts for approximately 90% of global humanoid robot exports and deeply integrates robotics into manufacturing, logistics and even consumer services.
This progress now worries US officials. Politico reports that Commerce Secretary Howard Lutnick recently warned company executives that Chinese robotics could be the next big national security problem. The administration’s concern is that subsidized Chinese manufacturers could dominate global robotics markets before American companies achieve similar scale. One of the meeting attendees openly expressed fear. “What we’ll end up with is an American brain with a Chinese body.”
These remarks reflect a broader concern prevailing in Washington. Technological leadership increasingly depends not only on invention but also on the ability to produce, distribute and commercialize technologies at scale. This is where China is gaining ground.
The war over chips turns into a war over ecosystems
The semiconductor industry remains at the center of competition as chips power everything from artificial intelligence systems to military technologies. This month, U.S. officials raised concerns about whether ASML’s cutting-edge ultraviolet lithography technology had somehow made its way to China. The Dutch company denied that such equipment was sent. The episode shows how chipmaking tools have become central to US strategy. Washington’s restrictions are designed not only to block certain products but also to restrict China’s ability to build advanced semiconductor ecosystems.
However, China continues to look for temporary solutions. Reuters reported that Chinese researchers have already developed a prototype EUV machine in an effort described as a Chinese version of the Manhattan Project. Although this effort is years behind ASML, it reflects its determination to eliminate the vulnerabilities created by foreign technology dependence. The bigger question is whether export controls slow China down faster than encouraging domestic innovation. Western analysts are increasingly unsure.
Beyond AI: Batteries, biotechnology and industrial power
One reason many American experts worry is that China’s advances are no longer limited to artificial intelligence or semiconductors. The New York Times recently argued that China is already moving into global leadership positions in sectors such as batteries, solar technology, rare earth elements and life sciences, pointing to CATL, the world’s largest battery maker, whose cutting-edge technology can reportedly deliver 250 miles of driving range on less than a ten-minute charge. The NYT report described the reversal of a decades-old pattern. For years, foreign companies brought advanced technologies to China. Today, American companies are increasingly turning to Chinese firms for cutting-edge products and manufacturing expertise.
Biotechnology may be an even more important example. Pfizer CEO Albert Bourla warned earlier this year that China’s scientific capabilities were growing rapidly. Speaking at a Council on Foreign Relations event, Bourla said US dominance in biotechnology was being challenged for the first time in decades. According to Bourla, China has spent years systematically investing in research institutions, talent development, intellectual property protection and regulatory reforms. Eight of the top ten institutions in the 2025 Nature Index are now Chinese. His conclusion was notable because it focused on American complacency rather than China’s strengths. He argued that the United States is paying too much attention to slowing down China and not enough attention to developing its own innovation ecosystem.
Winning artificial intelligence may not be enough
Perhaps the most important warning comes from Nobel laureate Simon Johnson and MIT’s Elisabeth Reynolds. In a recent Bloomberg article, they argue that America may win the AI race but still lose the broader technology competition. They note that China has spent two decades building industrial and innovation capacity across multiple sectors. According to research cited in the article, the United States was a leader in 61 of 64 pioneering technologies 20 years ago. Three years ago, China was leading in 57 of these 64 regions.
The authors point to drones, biotechnology, critical minerals, semiconductors, quantum computing and advanced manufacturing as areas where China is gaining significant advantages or narrowing America’s historic lead. China now produces 70% to 80% of global production in the field of drones alone. The lesson here is that technological power increasingly relies on entire ecosystems. Production capacity, supply chains, skilled workforce, financing and distribution are as important as scientific breakthroughs.
The gap continues but is gradually narrowing.
China could not surpass the USA in technology. American firms continue to dominate pioneering artificial intelligence models, advanced chip design, and many important research areas. The most advanced AI innovation still comes only from the United States. But focusing solely on who is ahead misses the more important trend. In every sector, China is demonstrating the ability to meet constraints, innovate under pressure, and build industrial power on a massive scale. Once dependent on Western technology, the country is increasingly producing technologies that Western companies want to use, study or compete with.
The LineShine supercomputer may not threaten America’s AI supremacy. But this is another reminder that the challenge facing China is no longer theoretical. The world’s second largest economy is no longer just trying to catch up. In many areas, it is already close enough to make America uneasy.
