Nvidia and AMD AI Chips to China Under New Revenue-Sharing Rules
Trade and technology are colliding once again as Nvidia and AMD AI chips to China become the subject of a new revenue-sharing deal with the United States. Under the arrangement, both companies can sell certain high-end AI chips to Chinese customers if they agree to give the U.S. government 15% of the revenue from those sales. This policy marks a shift from earlier national security-driven restrictions toward a more financially motivated approach, signaling how economic priorities are influencing tech exports in 2025.
Image Credits:Costfoto/NurPhoto / Getty Images
The agreement applies to Nvidia’s H20 AI chips and AMD’s MI308 chips, both designed with performance capabilities that make them critical in AI-driven applications. While the U.S. had previously imposed strict export controls on advanced AI chips to China, the recent licensing approval shows a nuanced strategy—allowing sales under strict conditions to balance economic interests with geopolitical considerations.
Why the US is Allowing Nvidia and AMD AI Chips to China
For the past few years, AI chip exports have been at the heart of U.S.-China technology tensions. Earlier restrictions aimed to limit China’s access to high-performance chips that could be used for advanced AI applications, particularly in sensitive areas such as surveillance, defense, and large-scale data analytics.
However, the landscape began to shift when Nvidia committed to significant U.S.-based investments, pledging up to $500 billion for data center infrastructure. This economic commitment, coupled with ongoing trade negotiations involving rare-earth materials vital for tech manufacturing, has influenced the decision to permit these sales.
U.S. Commerce Secretary Howard Lutnick indicated that these policy adjustments are tied to broader trade discussions, especially concerning rare-earth elements used in rechargeable batteries for electric vehicles and other high-tech products. By allowing these transactions with a revenue-sharing model, the government is attempting to safeguard national interests while maintaining a foothold in one of the world’s largest markets for AI hardware.
Economic Implications of Nvidia and AMD AI Chips to China
The decision to approve licenses for Nvidia and AMD AI chips to China could have far-reaching economic effects. For the companies, China represents a major market with growing demand for AI infrastructure, making the 15% revenue cut a cost they might be willing to accept for continued access.
Nvidia’s H20 chips and AMD’s MI308 models are tailored for high-performance computing, machine learning, and AI inference tasks. China’s appetite for such technology remains strong, driven by industries ranging from autonomous vehicles to cloud computing. Allowing these sales ensures that American companies remain competitive internationally, especially against rivals from other countries that face fewer export restrictions.
From the U.S. government’s perspective, the revenue-sharing model provides a steady income stream while maintaining oversight of technology exports. This could serve as a precedent for future deals involving other advanced technologies, creating a hybrid framework of regulation and monetization.
Balancing Security Concerns and Trade Interests
While the revenue-sharing arrangement might benefit both the companies and the U.S. treasury, it has not been without criticism. Some national security experts argue that enabling the sale of Nvidia and AMD AI chips to China could still pose risks if the technology is repurposed for military or surveillance applications. Former government officials have urged caution, emphasizing that even modified or limited-performance chips could contribute to strategic technological capabilities in China.
On the other hand, supporters of the deal believe that a controlled, licensed, and taxed export structure is better than a complete ban, which could push Chinese companies to seek alternatives from other suppliers, potentially hurting U.S. tech influence in the long run. They argue that staying in the market, even under constrained terms, ensures the U.S. retains a competitive edge and market intelligence.
The move also reflects a growing recognition that in the interconnected global tech ecosystem, outright bans may be less effective than strategic trade arrangements. As AI hardware continues to evolve, policies will need to strike a delicate balance between protecting national security, supporting economic growth, and fostering innovation.
Post a Comment