Tingbo He, president of Huawei Semiconductor, delivers a presentation at an industry conference in Shanghai on May 25, 2026.
Huawei
SHANGHAI – Chinese tech giant Huawei announced a new approach to developing advanced semiconductors on Monday, despite U.S. sanctions Nvidia is struggling to sell its high-end chips in China.
Huawei said in the fall it developed a new technical approach called “LogicFolding” to make its Kirin smartphone chips.
This breakthrough comes as Nvidia subject to US export restrictions in China and Apple is struggling with renewed competition from Huawei in the world’s second-largest consumer economy.
Huawei’s Mate 60 smartphone, released in 2023, featured 5G connectivity with an advanced chip that helped the company regain market share from Apple.
While U.S. restrictions have prevented Nvidia from selling its most advanced chips to China in recent years, Beijing is instead pushing to support homegrown technology. Last week, Nvidia CEO Jensen Huang told CNBC that the US chipmaker had “ceded” the Chinese market to Huawei.
“For Nvidia, this means the window of opportunity to sell advanced chips like the H200 in China is narrowing,” said George Chen, partner and co-chair of digital practice at The Asia Group.
“This development is likely to increase concerns in Washington, where Huawei remains a symbol of US export restrictions,” he said.
How Huawei rose from a telecommunications company to a Chinese “all-rounder” in the field of AI
Huawei said its new chip technology could deliver capabilities equivalent to 1.4-nanometer process technology by 2031 – while global chip leader TSMC has begun mass production of 2-nanometer chips.
Nanometer processes refer to chip manufacturing technology, with smaller nodes typically enabling faster and more efficient semiconductors.
Paul Triolo, head of technology for Asia and the Americas at DGA Group, was skeptical of Huawei’s 1.4-nanometer claim.
“A stacked/folded design can result in effective density gains, but that does not mean Huawei has solved the entire process, yield, power, thermal and device performance issues associated with true 1.4nm-class manufacturing,” he said.
It is not possible to evaluate advanced extreme ultraviolet (EUV) lithography machines from the Dutch chip equipment manufacturer ASML“Huawei has been forced to look for alternatives to chip development to remain competitive in AI,” said Neil Shah, vice president of research at Counterpoint Research.
“However, this parallel semiconductor path has not yet been proven at large scale. This approach can impose severe thermal limitations and packaging complexity, which can negatively impact production yield,” said Shah.
Huawei’s efforts to deploy the technology in its flagship Mate 90 series of smartphones this fall would be a technical feat, but scaling it to AI data centers would serve as the “ultimate litmus test for China’s creative solution to Western sanctions,” he added.
Academic ambitions
Huawei is also seeking greater academic recognition of its semiconductor research. On Monday, the company described its findings as “Law of Tau” or “τ scaling” and claimed it addresses the challenges facing the semiconductor industry.
Huawei said it has developed and mass-produced 381 chips based on the “τ Scaling Law” over the past six years.
Semiconductor development has for decades relied on “Moore’s Law,” an observation that states the number of transistors doubles about every two years—providing more computing power while reducing costs. But even Nvidia’s Huang has stated that Moore’s Law no longer applies to future chip development.
“Huawei essentially turns a technical strategy into a ‘law,'” Triolo said.
The new principle “is more of a system-level optimization doctrine: shorten lines, stack logic, improve memory semantics and co-design chips, packages, software and clusters,” he said.
Still, challenges remain around thermal management and large-scale manufacturing, Triolo said.
According to Tingbo He, president of Huawei’s semiconductor business, Huawei’s new chip architecture expands the layout from one to two layers, significantly increasing power efficiency.
This structure allows transistors to interact with each other at more points, He, who is also director of the company’s science committee, said at the Institute of Electrical and Electronics Engineers’ International Symposium on Circuits and Systems.
But she acknowledged that challenges remain as Huawei is just beginning its decades-long development of the new technology.
China’s AI chip companies are quickly going public. But its crown jewel remains private. Choose CNBC as your preferred source on Google and never miss a moment from the most trusted name in business news.