Will NVIDIA and AMD Pass the 15% Fee to Chinese Buyers?

Will NVIDIA and AMD Pass the 15% Fee to Chinese Buyers?

Analyst: FuturumAI
Publication Date: August 11, 2025

The recent 15% export license fee on NVIDIA’s H20 and AMD’s MI308 to China represents a significant policy development in the global AI semiconductor market.

What You Need to Know:

  • The Trump administration imposed a 15% fee on licensed exports of NVIDIA’s H20 and AMD’s MI308 AI chips to China.
  • NVIDIA posted $26.0B revenue and $6.12 EPS last quarter, driven by AI data center demand.
  • AMD reported $5.84B revenue and $0.69 EPS last quarter, with strong MI300 adoption.
  • Due to high demand, Futurum Equities expects NVIDIA to pass the fee to Chinese buyers; AMD may also achieve pass-through.
  • Policy changes may alter allocation and delivery timelines for global customers.

Recommendations:

To navigate the new policy environment, AI chip buyers and investors should:

  • Monitor allocation strategies as China re-enters the demand pool under the license framework.
  • Plan procurement timelines to account for possible shifts in delivery schedules.
  • Track management commentary on fee pass-through and licensing throughput as key variables for margin impact.

Will NVIDIA and AMD Pass the 15% Fee to Chinese Buyers?

Analyst Take: The recent 15% export license fee on NVIDIA’s H20 and AMD’s MI308 to China represents a significant policy development in the global AI semiconductor market.

While it adds a cost layer, demand for NVIDIA’s AI accelerators in China is so intense that we expect the fee to be fully passed on to buyers without dampening sales.

AMD’s growing credibility with the MI308 suggests it may also succeed in passing along most of the cost.

Both companies stand to benefit from improved capacity planning and global allocation strategies as licensing becomes the new norm.

Bullish factors include strong AI demand in China and pricing power for both NVIDIA and AMD. Risks include potential retaliatory measures and the complexity of license processing.

For more information, visit the NVIDIA newsroom here or AMD’s newsroom here.

Declaration of generative AI and AI-assisted technologies in the writing process: This content has been generated with the support of artificial intelligence technologies. Due to the fast pace of content creation and the continuous evolution of data and information, The Futurum Group and its analysts strive to ensure the accuracy and factual integrity of the information presented. However, the opinions and interpretations expressed in this content reflect those of the individual author/analyst. The Futurum Group makes no guarantees regarding the completeness, accuracy, or reliability of any information contained herein. Readers are encouraged to verify facts independently and consult relevant sources for further clarification.

Disclosure: Futurum is a research and advisory firm that engages or has engaged in research, analysis, and advisory services with many technology companies, including those mentioned in this article. The author does not hold any equity positions with any company mentioned in this article.

Analysis and opinions expressed herein are specific to the analyst individually and data and other information that might have been provided for validation, not those of Futurum as a whole.

Other Insights from Futurum:

The Six Five Pod | EP 270: Silicon Strategies: Intel’s 14A Gambit and the Future of Chip Manufacturing

AMD Q2 FY 2025 Sales Beat Offset by MI308-Linked EPS Decline

NVIDIA’s GB300 NVL72 Power Shelf: Answer to AI’s Grid Strain?

Author Information

FuturumAI

This content is written by a commercial general-purpose language model (LLM) along with the Futurum Intelligence Platform, and has not been curated or reviewed by editors. Due to the inherent limitations in using AI tools, please consider the probability of error. The accuracy, completeness, or timeliness of this content cannot be guaranteed. It is generated on the date indicated at the top of the page, based on the content available, and it may be automatically updated as new content becomes available. The content does not consider any other information or perform any independent analysis.

Related Insights
Agentic Workloads Reshape
June 17, 2026

How will Qualcomm’s AI Bet Solve for NVIDIA’s Data Center Gaps as Agentic Workloads Reshape the Chip Market?

Olivier Blanchard, Research Director & Practice Lead, Intelligent Devices at Futurum, on Qualcomm's Investor Day, and whether Qualcomm can challenge NVIDIA's data center dominance....
How Desktop AI Hubs Could Deflect Over 56.23 TWh of Industrial Data Center Load by 2035
June 12, 2026

How Desktop AI Hubs Could Deflect Over 56.23 TWh of Industrial Data Center Load by 2035

Olivier Blanchard and Brendan Burke, Research Directors at Futurum, share their insights on how high-performance small-form-factor desktop AI PCs such as the DGX Spark and Mac Mini could form the...
AWS Graviton5 Reframes the CPU as Agentic AI Infrastructure
June 12, 2026

AWS Graviton5 Reframes the CPU as Agentic AI Infrastructure

Brendan Burke, Research Director at Futurum, analyzes how AWS Graviton5's general availability redefines CPU architecture for agentic AI, with Meta deploying tens of millions of cores and customers halving their...
Cadence and Synopsys Accelerate Agentic EDA Race at Computex
June 11, 2026

Cadence and Synopsys Accelerate Agentic EDA Race at Computex

Brendan Burke, Research Director at Futurum, assesses how Cadence and Synopsys are accelerating the agentic EDA race, with Cadence reaching Level 5 autonomy and Synopsys expanding into multi-physics workflows....
Can AMD EPYC Extend Its Lead Over Vera and Xeon in the Agentic Data Center?
June 11, 2026

Can AMD EPYC Extend Its Lead Over Vera and Xeon in the Agentic Data Center?

Brendan Burke, Research Director at Futurum, examines AMD's EPYC 9965's claimed 2.37x NVIDIA Vera throughput and 1.6x Intel Xeon performance in 100 kW rack-scale agentic workloads....
Canonical’s Ubuntu TPU Optimization Shows the Coming Structural Shift in Enterprise AI Infrastructure
June 11, 2026

Canonical’s Ubuntu TPU Optimization Shows the Coming Structural Shift in Enterprise AI Infrastructure

Guy Currier at Futurum examines Canonical’s launch of optimized Ubuntu images for Google Cloud TPU virtual machines and its strategic implications for enterprise AI infrastructure economics, accelerator diversification beyond GPUs,...

Book a Demo

Newsletter Sign-up Form

Get important insights straight to your inbox, receive first looks at eBooks, exclusive event invitations, custom content, and more. We promise not to spam you or sell your name to anyone. You can always unsubscribe at any time.

All fields are required






Thank you, we received your request, a member of our team will be in contact with you.