Friday, 1 May 2026
Subscribe
logo
  • AI Compute
  • Infrastructure
  • Power & Cooling
  • Security
  • Colocation
  • Cloud Computing
  • More
    • Sustainability
    • Industry News
    • About Data Center News
    • Terms & Conditions
Font ResizerAa
Data Center NewsData Center News
Search
  • AI Compute
  • Infrastructure
  • Power & Cooling
  • Security
  • Colocation
  • Cloud Computing
  • More
    • Sustainability
    • Industry News
    • About Data Center News
    • Terms & Conditions
Have an existing account? Sign In
Follow US
© 2022 Foxiz News Network. Ruby Design Company. All Rights Reserved.
Data Center News > Blog > Cloud Computing > Nvidia Sees $5.5B Hit From New Trump China Curbs on Chips
Cloud Computing

Nvidia Sees $5.5B Hit From New Trump China Curbs on Chips

Last updated: April 17, 2025 10:29 pm
Published April 17, 2025
Share
Nvidia Sees $5.5B Hit From New Trump China Curbs on Chips
SHARE

(Bloomberg) — President Donald Trump’s administration has barred Nvidia Company from promoting its H20 chip in China, an escalation of Washington’s tech battle with Beijing that can value the corporate billions of {dollars} and hamstring a product line it explicitly designed to adjust to earlier US curbs. 

The US authorities knowledgeable Nvidia on Monday that the H20 would require a license to export to China “for the indefinite future.” The brand new guidelines deal with Washington’s considerations that “the lined merchandise could also be utilized in, or diverted to, a supercomputer in China,” the corporate stated in a submitting. Nvidia warned it’ll report about $5.5 billion in writedowns in the course of the present quarter, tied to stock and commitments for the chip.

Its shares slid 6.9% in New York, main a broad selloff in semiconductor companies from the US to Japan. South Korean chipmakers Samsung Electronics Firm and SK Hynix had been down by round 3%. Superior Micro Gadgets, which additionally stated it expects to take a cost associated to the Trump chip curbs, slumped as effectively. 

The slide was compounded as ASML Holding reported disappointing monetary outcomes for the primary quarter. The Dutch firm, which sells essentially the most superior machines to fabricate chips, stated its orders for the primary quarter had been virtually a billion euros lower than anticipated and warned that tariffs will add to uncertainty within the business.

Associated:Google Launches Ironwood TPU For Subsequent-Gen AI Inference

AMD stated it expects to take a cost of as a lot as $800 million after finishing an preliminary evaluation of the Trump administration’s new license requirement for the export of the chipmaker’s MI308 merchandise. 

See also  Nvidia CEO Jensen Huang sings praises of processor in Nintendo Switch 2

“The Commerce Division is issuing new export licensing necessities on the Nvidia H20, AMD MI308 and their equivalents,” a spokesperson for the division stated in an announcement. “The Commerce Division is dedicated to performing on the president’s directive to safeguard our nationwide and financial safety.”

The brand new restrictions are among the many clearest indicators but that Trump will keep the course on Washington’s efforts to restrict China’s semiconductor and synthetic intelligence ambitions. His staff has been exploring such a step for the reason that begin of the administration, Bloomberg Information reported in January, persevering with conversations that started beneath President Joe Biden. 

Trump’s commerce secretary, Howard Lutnick, has pledged to be “very robust” on China chip curbs – notably after the emergence of AI startup DeepSeek. He not too long ago sanctioned dozens of Chinese language companies that Trump officers allege are aiding Beijing’s army efforts. Within the Asian nation’s chipmaking enviornment, Huawei Applied sciences Co.’s AI accelerators are the closest competitor to Nvidia’s and AMD’s choices, although they nonetheless lag considerably when it comes to efficiency.

Associated:Nvidia, Cassava’s AI Manufacturing facility in Africa Tie-As much as Value $720M

“The US authorities’s strengthened rules on Nvidia might be seen as a part of the continued commerce warfare between the US and China,” stated Tomo Kinoshita, world market strategist at Invesco Asset Administration. “Nonetheless, it is usually a coverage pushed by considerations over China’s rise within the electronics sector, and in that sense, it’s prone to grow to be a everlasting coverage.”

US officers first barred Nvidia and different AI chipmakers from promoting their most superior fashions to China in October 2022, over considerations that the expertise might give Beijing a army edge. Since then, the China controls have ballooned to incorporate an more and more giant set of semiconductor manufacturing instruments, in addition to a wider vary of each processors and high-bandwidth reminiscence chips, that are important for AI functions.

See also  The Intersection of technology and human behaviour in cybersecurity

Nvidia designed the H20 chip in response to 2023 measures that restricted one other China-focused processor, the H800, which the corporate debuted following the preliminary curbs. The H20 is meant to be much less highly effective, and it doesn’t carry out in addition to Nvidia’s top-of-the-line choices for the coaching of fashions. However it will possibly nonetheless be used to develop and run AI software program and providers, and it’s notably well-suited to the inference stage: the purpose at which AI fashions acknowledge patterns and draw conclusions. 

Associated:SoftBank Seals $6.5B Deal for Chip Designer Ampere

Below the Biden administration, officers had ready rules that may have restricted Nvidia’s H20 and B20 chips, a senior Biden official stated, they usually offered these guidelines to then-Commerce Secretary Gina Raimondo. Biden workers estimated such curbs would make it round 3% to six% costlier to develop an AI mannequin in China – at a roughly $12 billion value to Nvidia, stated the official, who requested anonymity to talk candidly.

Requested how Biden’s staff arrived that quantity, the official stated it was based mostly on an estimate that Nvidia would promote round one million H20s in China, at $12,000 a bit. Nvidia declined to remark.

However Raimondo in the end selected to not pursue the measures, partly as a result of the Biden administration was engaged on a separate set of AI chip restrictions affecting a lot of the world. That so-called AI diffusion rule, unveiled throughout Biden’s ultimate week in workplace, is now the topic of a fierce lobbying effort in Washington. 

See also  Goldman Sachs sees AI investment shift to data centres

Nvidia has staunchly opposed the framework, arguing that it’ll weaken American corporations and reinforce China’s dedication to make itself impartial of US expertise. Trump’s staff has indicated they wish to strengthen and streamline the measures, Bloomberg Information has reported, although what that entails stays unclear. Corporations are required to start out complying with the worldwide restrictions in a few month. 

As for the H20 guidelines, Nvidia’s writedown means that the corporate might miss out on $14 billion to $18 billion in income for the 12 months, Bloomberg Intelligence analysts Kunjan Sobhani and Oscar Hernandez Tejada stated in a observe. “If restrictions persist, Nvidia’s data-center publicity to China might normalize to low- to mid-single digits, just like early 2024 ranges,” they stated. That was earlier than manufacturing of the H20 ramped up. 

The brand new restrictions comply with an NPR report that Trump had backed off from pursuing the H20 controls, in alternate for Nvidia investing in information facilities. The corporate simply introduced that it could construct as much as $500 billion price of AI infrastructure within the US over the subsequent 4 years, a determine that features plans that had been already underway.



Source link

TAGGED: 5.5B, China, Chips, curbs, Hit, Nvidia, sees, Trump
Share This Article
Twitter Email Copy Link Print
Previous Article Source: Rubaitul Azad/Unsplash Huawei’s AI hardware breakthrough challenges Nvidia’s dominance
Next Article nLighten and Shell Spain partner nLighten and Shell Spain partner
Leave a comment

Leave a Reply Cancel reply

Your email address will not be published. Required fields are marked *

Your Trusted Source for Accurate and Timely Updates!

Our commitment to accuracy, impartiality, and delivering breaking news as it happens has earned us the trust of a vast audience. Stay ahead with real-time updates on the latest events, trends.
FacebookLike
TwitterFollow
InstagramFollow
YoutubeSubscribe
LinkedInFollow
MediumFollow
- Advertisement -
Ad image

Popular Posts

Partnership Targets Quantum-Accelerated AI Factories

Quantum computing firm Orca Computing within the U.Okay. and ParTec, a German high-performance computing and…

May 19, 2025

Pulsant optimises Reading SE3 data centre connectivity to drive regional AI ambitions

Digital infrastructure chief Pulsant right now introduced upgrades to its Studying SE3 information centre community…

March 20, 2025

Researchers improved AI agent performance on unfamiliar tasks using ‘Dungeons and Dragons’

Be part of our every day and weekly newsletters for the newest updates and unique…

January 12, 2025

atNorth joins UN Global Compact

atNorth, the main Nordic colocation, high-performance computing, and synthetic intelligence service supplier, has grow to…

February 4, 2025

Endace releases OSm 7.3 network packet capture update

As community safety threats improve and laws similar to DORA and GDPR place larger calls…

January 23, 2026

You Might Also Like

The role of AI in enterprise infrastructure operations
Cloud Computing

The role of AI in enterprise infrastructure operations

By saad
Keppel starts work on floating data centre in Singapore
Cloud Computing

Keppel starts work on floating data centre in Singapore

By saad
The last piece in the DC construction puzzle: Ongoing operations
Cloud Computing

The last piece in the DC construction puzzle: Ongoing operations

By saad
Google Cloud and NVIDIA logos as, at the Google Cloud Next conference, the companies outlined their hardware roadmap designed to address the cost of AI inference at scale.
AI & Compute

NVIDIA and Google infrastructure cuts AI inference costs

By saad

About Us

Data Center News is your dedicated source for data center infrastructure, AI compute, cloud, and industry news.

Top Categories

  • AI & Compute
  • Cloud Computing
  • Power & Cooling
  • Colocation
  • Security
  • Infrastructure
  • Sustainability
  • Industry News

Useful Links

  • Home
  • Contact
  • Privacy Policy
  • Terms & Conditions

Find Us on Socials

© 2026 Data Center News. All Rights Reserved.

© 2026 Data Center News. All Rights Reserved.
Welcome Back!

Sign in to your account

Lost your password?
We use cookies to ensure that we give you the best experience on our website. If you continue to use this site we will assume that you are happy with it.
You can revoke your consent any time using the Revoke consent button.