FREE REPORT: A New Way to Earn Passive Income in 2025 DOWNLOAD

Huawei CEO says firm is finding workarounds to navigate chip gap with US

In this post:

  • Huawei uses cluster computing, novel chip designs to bridge performance gaps.
  • The company now invests ¥180 billion annually in theoretical and product research development.
  • Its Ascend AI chips in clusters help narrow the gap with Nvidia’s.

Huawei’s chipmaking remains a generation behind US rivals, but the company is deploying creative workarounds to close the gap, according to remarks by CEO Ren Zhengfei published this week by local media.

Despite stringent American export curbs imposed since 2019, Huawei is leaning on techniques such as cluster computing and innovative multi-element “compound” chips to boost performance, Zhengfei told China’s official Communist Party mouthpiece.

Huawei banks on innovation to narrow the gap

Zhengfei emphasized that Huawei dedicates roughly 180 billion yuan (about US$25 billion) each year to R&D, splitting that investment between theoretical exploration and product development.

“Without a solid theoretical foundation, breakthroughs are impossible, and without breakthroughs, we cannot hope to narrow the lead established by US firms,” he said, noting that approximately one-third of the annual research budget underpins fundamental science, with the remaining two-thirds channeled into practical engineering and design.

At the heart of Huawei’s strategy lies a trio of complementary approaches. First, the firm pairs advanced mathematics with conventional physics to simulate and optimize chip behaviour, sidestepping some of the constraints of Moore’s Law, the long-held principle that transistor counts double roughly every two years.

Second, by blending varied semiconductor materials into compound chips, Huawei seeks to eke out new performance gains beyond what single-material devices can achieve.

Finally, the company is harnessing large-scale cluster computing, linking hundreds of individual processors into a unified system, to compensate for any shortfall in the raw speed of a single chip.

“Through weaving mathematics into physical design and augmenting Moore’s Law with non-Moore’s Law techniques, then layering on cluster computing, we can attain results suitable for real-world applications.”

Zhengfei.

He added that software has not proven to be a bottleneck for Huawei; rather, the limitations lie in the hardware itself.

See also  Getty Images vs. Stability AI copyright lawsuit begins in the UK

Zhengfei’s frank admission of his firm’s lag comes as senior US and Chinese trade negotiators continue discussions in London, where America’s tech restrictions on China are high on the agenda.

The company wants to prove a point

Since 2019, Washington has barred Chinese firms, including Huawei, from obtaining the most advanced processors and the high-precision equipment needed to manufacture them, in a bid to curb Beijing’s technological and military ambitions.

Until now, neither Zhengfei nor any other Huawei executive has publicly detailed the company’s chipmaking efforts. His comments mark the first official peek at how the Chinese firm is responding to these challenges.

“The United States has perhaps overstated Huawei’s prowess,” Zhengfei acknowledged.

“We are not as extraordinary as some claim, we must work diligently to prove ourselves.”

Zhengfei

In parallel with its defensive measures, Huawei has been actively promoting its Ascend family of artificial-intelligence chips, which vie with products from Nvidia, the undisputed leader in AI processors. In April, Huawei unveiled its “AI CloudMatrix 384” system, which stitches together 384 Ascend 910C chips in a single cluster designed for AI model training.

Industry analysts have pointed out that on certain benchmarks, this configuration can even outpace Nvidia’s GB200 NVL72 setup, though the American firm still dominates many other metrics.

See also  A frustrated Mark Zuckerberg is personally hiring a "superintelligence" AI team at Meta

Nevertheless, the US Commerce Department recently ruled that deploying Ascend chips would breach existing export controls, underscoring the fraught intersection of global competition and national security.

Nvidia, for its part, has been unable to sell its most cutting-edge GPUs to Chinese buyers, a situation that has enabled Huawei to regain some market share domestically, even as its chips remain technically a step behind.

Looking ahead, Zhengfei expressed optimism. He highlighted the promise of compound chips, devices constructed from multiple semiconductor materials, as a potential game-changer. And he reiterated his confidence that, through persistent investment in both theory and engineering, Huawei will continue shrinking the performance gap with its US counterparts.

“Worry over Huawei’s chip supply is unnecessary,” he concluded, urging stakeholders to recognize the firm’s resilience and ingenuity under pressure.

KEY Difference Wire helps crypto brands break through and dominate headlines fast

Share link:

Disclaimer. The information provided is not trading advice. Cryptopolitan.com holds no liability for any investments made based on the information provided on this page. We strongly recommend independent research and/or consultation with a qualified professional before making any investment decisions.

Most read

Loading Most Read articles...

Stay on top of crypto news, get daily updates in your inbox

Editor's choice

Loading Editor's Choice articles...

- The Crypto newsletter that keeps you ahead -

Markets move fast.

We move faster.

Subscribe to Cryptopolitan Daily and get timely, sharp, and relevant crypto insights straight to your inbox.

Join now and
never miss a move.

Get in. Get the facts.
Get ahead.

Subscribe to CryptoPolitan