Huawei Captures China AI Chip Market as U.S. Prepares DUV Equipment Crackdown
Huawei's AI processor sales are projected to hit $12 billion this year, a 60% jump, as China builds a domestic stack from silicon to services. Washington is drafting controls on older lithography tools to slow the momentum.

China has assembled what analysts now describe as a functioning AI self-reliance system, spanning semiconductors through cloud services, as Huawei Technologies emerges as the dominant domestic chip supplier and U.S. lawmakers prepare a new wave of export restrictions targeting older manufacturing equipment.
Huawei expects AI chip sales to reach approximately $12 billion in 2026, up more than 60 percent from $7.5 billion in 2025, according to projections cited by the Financial Times. Orders for the company's Ascend 950PR processor, which entered mass production in March, have surged as Chinese tech firms seek alternatives to Nvidia hardware blocked by existing U.S. controls. Morgan Stanley forecasts China's AI chip market will reach $51 billion by 2030, with domestic suppliers capturing 76 percent of that total.
The momentum has prompted a legislative response in Washington. The U.S. House of Representatives has introduced the Multilateral Agreement on Technology Controls for Hardware, known as the MATCH Act, which would expand restrictions beyond extreme ultraviolet lithography equipment to include deep ultraviolet systems—older tools that Chinese fabs still rely on for production. The bill targets all facilities, subsidiaries, and affiliates of major Chinese semiconductor companies. Separately, the State Department is preparing countermeasures against "distillation" techniques, in which Chinese developers train models using outputs from U.S. systems.
DeepSeek's V4 model preview, released last month, demonstrated improved compatibility with Huawei chips, a shift analysts say will accelerate commercialization of domestic AI computing. BOC International stated that the model's performance on locally produced silicon marks a turning point for China's AI stack. The development sent ripples through parts of the U.S. industry earlier this year, though observers note persistent constraints below the model layer.
"Today, in the United States, you have the data, you have the computing access, you have the chips, you have the talent. China does a very good job on the top of the stack, but is lacking some elements below," said Christophe Fouquet, CEO of ASML, the Dutch firm that holds a monopoly on EUV lithography machines, speaking at a TechCrunch event. Without access to EUV technology, Chinese chipmakers cannot manufacture the most advanced semiconductors, and models built on older hardware operate at a compounding disadvantage regardless of software quality.
(China's government has pledged at least 7 percent annual growth in nationwide research and development spending through 2030 under its five-year plan, with an "AI plus" blueprint outlining integration across healthcare, education, and judicial systems. Shenzhen courts reported processing 50 percent more cases last year, partly aided by AI tools.)
Samm Sacks, a senior fellow at New America focused on Chinese technology policy, noted that export controls have slowed chipmaking capabilities and remain "the Achilles' heel of many AI labs that need advanced AI chips." But the restrictions have also driven tighter coordination across China's design, manufacturing, and adoption chain. "Over time this dynamic could fuel, not foil, China's ambitions," Sacks said.
The rivalry reflects a broader decoupling in semiconductor supply chains. U.S. chip stocks posted historic gains in recent weeks, with Intel recording its best single-day performance since 1987 and Nvidia's market capitalization breaching $5 trillion. Bruce Bateman, chief analyst at Omdia, described the rally as "winning streaks not seen since the 1970s." Meanwhile, Samsung Electronics reported an eightfold increase in first-quarter operating profits, driven by explosive chip business growth, and the Pentagon confirmed it is expanding use of Google's Gemini AI model after designating Anthropic a supply chain risk.
The divergence underscores a strategic split: U.S. firms are betting on continued infrastructure spending and access to cutting-edge lithography, while China is optimizing software and system integration to extract maximum performance from constrained hardware. Whether the MATCH Act can arrest China's progress without disrupting allied semiconductor industries remains an open question, as does the timeline for any multilateral enforcement mechanism the bill envisions.
Keywords
Sources
https://www.chosun.com/english/industry-en/2026/05/07/34RF4XBQMBAOXD54MSKZGHJE3E/
Frames U.S. legislative push to expand controls to DUV equipment and distillation techniques as escalation in semiconductor rivalry.
https://www.chosun.com/english/industry-en/2026/05/06/FUH44AHFBVE47H4BIJG6VGDS74/
Emphasizes Huawei's 60% sales growth and DeepSeek V4's compatibility with domestic chips as proof of functional self-reliance system.
https://techcrunch.com/2026/05/06/five-architects-of-the-ai-economy-explain-where-the-wheels-are-coming-off/
ASML CEO highlights China's strength at software layer but persistent disadvantage in advanced semiconductor manufacturing without EUV access.
https://www.latimes.com/business/story/2026-05-06/u-s-china-ai-gap-has-closed-and-silicon-valley-is-starting-to-notice
Explores how export controls have paradoxically improved coordination across China's tech supply chain, potentially fueling long-term ambitions.
