By Lingling Wei

In 2019, while researching a book on the U.S.-China competition, I visited Huawei's sprawling new R&D campus in Dongguan, a city about an hour's drive from the company's Shenzhen headquarters. The place was something to behold: 300 acres, more than 25,000 employees, and 12 European-style "towns" -- Paris, Luxembourg, and others -- connected by a miniature train modeled on the Shenzhen metro. A Huawei executive explained the design to me with a straight face: "We're doing well in Europe."

The campus told you everything about the company's confidence. In a Wall Street Journal interview that November, founder Ren Zhengfei invited President Trump to come visit and brushed off U.S. sanctions with characteristic bravado. "They may as well keep us there forever," he said, referring to the sanctions. "We'll be fine without them."

More than six years later, Huawei is still here -- and still projecting confidence. Last week, its semiconductor chief He Tingbo took the stage at a chip conference in Shanghai and announced that Moore's Law had a successor. The principle that chips roughly double in power every two years has been the industry's North Star since the 1960s. He (pronounced "huh"), known as Huawei's "chip queen," called it the Tau Law.

But a careful reading of the company's technical paper reveals something rather different from what Huawei intended.

The engineering is real. Think of a traditional chip like a single-story building -- for decades, the way to pack more onto it was to shrink everything inside. Huawei's technique takes a different approach: instead of building smaller, it builds upward, stacking two layers of circuitry on top of each other so signals travel shorter distances. The company said the approach would make its next chip 55% denser than its predecessor and could deliver cutting-edge performance by 2031.

"The engineering is genuinely impressive," noted Jimmy Goodrich, an independent semiconductor analyst who reviewed the paper. "The breakthrough framing is not." That's because Huawei, he said, is doing clever things with the tools available to them -- and that is the real story.

Building upward isn't a new idea -- TSMC, Intel, AMD and Samsung are all doing it too. Goodrich said the difference is that all of their designs are stacked atop chips made using a technology called extreme ultraviolet lithography, or EUV: the manufacturing process behind the industry's biggest advances over the past decade. EUV machines are made almost exclusively by a Dutch company called ASML, and U.S.-led export controls have barred their sale to China since 2019.

China doesn't have a homegrown alternative. Huawei is pitching stacking as a substitute for conventional chip advances rather than a complement to them because the key technology that would make it a complement is the one thing it can't get.

Huawei's paper on the Tau Law essentially admits this. It opens by noting that for companies whose "access to the most advanced lithography is constrained, the constraint became binding earlier and bears down more severely." A few pages later: "Assuming that another node would resolve the problem was no longer tenable."

Goodrich called it "the clearest statement yet from inside Huawei that they have accepted they can't break through the EUV barrier on any meaningful near-term horizon."

Now look at the gap. Huawei's goal is to match the performance of a cutting-edge chip by 2031, by stacking on older silicon. Goodrich noted that TSMC will reach that same performance level in 2028 -- and will be a generation beyond it by 2031, applying the same stacking techniques on far more advanced chips. "The realistic gap in 2031 is six to eight years," the analyst said, "not the three years" implied by Huawei's announcement.

There's one more problem, Goodrich said, and it's a practical one. Huawei's factories are estimated to produce usable chips only about 20% of the time, Goodrich said -- meaning four out of every five come out defective. Stacking requires bonding two chips together with extreme precision. If your yield is already shaky on one chip, it gets a lot shakier when you need two to come out right.

Huawei's announcement is likely to be heard by two audiences at once. To Washington, it says export controls aren't working -- useful cover for anyone looking to ease them. To Beijing, it justifies turning down the higher-quality Nvidia chip licenses that President Trump recently authorized but that China hasn't yet bought: we don't need American technology, we've built our own path. In other words, you might as well scrap all your export controls.

However, neither argument holds up against the paper's own contents. The export controls have worked -- but there are many loopholes. Case in point: On Sunday, the Bureau of Industry and Security, the agency overseeing such restrictions, issued guidance clarifying that a license requirement had always applied to offshore operations of Chinese companies -- an implicit acknowledgment that the rule had been circumvented by China-headquartered companies to smuggle advanced chips through their overseas operations.

Huawei is innovating around the controls -- impressively, in places -- but it's working around a wall it has now admitted, in its own writing, it can't climb over.

Back in 2019, standing on that Dongguan campus with white marble horses rearing in front of a French neoclassical facade, Huawei's defiance felt like a genuine statement of strength. The Tau Law announcement reads more like a statement of constraint, as Goodrich and others have pointed out -- dressed up, for political reasons, to look like something else.

Is China catching up in chips faster than Washington thinks -- or slower than Beijing claims? Write to me at lingling.wei@wsj.com. Include your full name and location, and I might publish your response in a coming issue (if you're reading this in your inbox, you can just hit reply).

This is an edition of the WSJ China newsletter, a weekly dispatch of exclusive insights on the contest between the U.S. and China, brought to you by the WSJ's top China correspondent. If you're not subscribed, sign up here.

China in a Few Headlines

   -- China is encouraging companies to adopt artificial intelligence while 
      pressuring them to avoid worker layoffs. 
 
   -- China memory-chip maker CXMT clears Shanghai listing review. 
 
   -- PDD Holdings, the owner of Temu, reported a profit miss driven by fierce 
      competition in China's e-commerce market. 
 
   -- China is exporting its factories worldwide, triggering fierce resistance 
      from global competitors. 

A Closer Look

The United Nations is facing a severe liquidity crisis and a projected "race to bankruptcy" by mid-August, driven largely by significant financial shortfalls from member states, including over $4 billion in arrears from the U.S. and $455 million from China. In response to this crunch, the U.N. has implemented historically large spending cuts, eliminated 3,000 secretariat posts, and slashed peacekeeping expenditures.

Reader Responses

Last week, we asked whether U.S. policy toward China will look meaningfully different after Trump's term ends -- or whether the underlying dynamics are now bigger than any one administration. Readers shared their thoughts:

"These underlying dynamics will persist across administrations. The Taiwan Strait, China's leverage over critical minerals, the race to lead in artificial intelligence and semiconductors, and broader questions of economic and military competition will not disappear when the Trump administration ends. Some of these challenges long predate Trump, while newer issues, such as AI, will almost certainly continue well into the future. That said, how future administrations continue to engage with these issues with China will likely be meaningfully different. Any renewed emphasis on alliances, foreign aid, multilateral institutions, or more predictable engagement with partners would represent a vastly different U.S. policy toward China than Trump's, even if the underlying dynamics remain." -- Isaiah Barber, Washington, D.C.

"It depends on the next administration. If Republicans win, I can see policy continuing albeit in a slightly less audacious manner. If Democrats win, I see appeasement given their recent embrace of socialism. They want to be able to say 'look at us, we are like you, let's get along." -- Bill Lechner, New York

"A change in the U.S. administration will most likely have no change in U.S. policy toward China regarding Taiwan. Strategic ambiguity has been in force during Democrat and Republican presidents and has worked well enough to be retained by both. China should also want to retain the peaceful status quo regarding Taiwan as any change in policy could result in some form of political or economic instability. Given the state of China's economy, they may not be able to continue their aggressive expansion beyond their borders. America's efforts to contain China will most likely be focused on tariffs, impacting China's export economy and foreign influence." -- Eugene P. Grace, Pennsylvania

(Responses have been condensed and edited.)

A Name in the News

Cheng Li-wun, chairwoman of Taiwan's Kuomintang, is set to begin a two-week U.S. tour that is billed as a peace mission. Her message aligns with Beijing's preferred framing, delivered at a moment of American ambiguity about its security commitments to Taipei.

About Us

WSJ China is a weekly newsletter with exclusive insights on the contest between the U.S. and China, brought to you by WSJ Chief China Correspondent Lingling Wei, with help from Zhao Yueling. Reach Lingling at lingling.wei@wsj.com or at @Lingling_Wei on X (if you're reading this in your inbox, you can just hit reply). Sign up to get an alert every time she publishes an article. Got a tip for us? Here's how to submit.

(END) Dow Jones Newswires

June 02, 2026 06:55 ET (10:55 GMT)

Copyright (c) 2026 Dow Jones & Company, Inc.