In a groundbreaking and highly unusual move, the U.S. government has forged a new agreement with semiconductor titans NVIDIA Corp. (NASDAQ: NVDA) and Advanced Micro Devices Inc. (NASDAQ: AMD), mandating that they remit 15% of their revenues derived from certain artificial intelligence (AI) chip sales to China. This unprecedented revenue-sharing model, confirmed by President Donald Trump on August 11, 2025, marks a significant departure from traditional export control policies and immediately casts a shadow over the companies' profit margins and stock performance, even as it reopens a critical market.
The deal, which specifically targets less powerful versions of their flagship AI processors—NVIDIA's H20 AI accelerator and AMD's MI308 chips—aims to balance national security concerns with the economic interests of American technology firms. While it allows these companies to re-enter the lucrative Chinese market after a period of tightened restrictions, the direct cut to their top line has sparked investor apprehension, leading to slight declines in their stock prices despite the re-issuance of export licenses.
An Unprecedented Accord: What Happened and Why It Matters
The genesis of this extraordinary agreement lies in the U.S. government's ongoing efforts to curb China's access to advanced AI semiconductor technology, citing national security concerns. In April 2025, the Trump administration had initially halted sales of NVIDIA's H20 and AMD's MI308 chips to China, which were specifically designed to comply with previous export restrictions. This move intensified the tech trade war and threatened to significantly impact the revenue streams of both companies.
However, a reversal of this decision in July paved the way for the current revenue-sharing condition. President Trump revealed that initial negotiations sought a 20% revenue share, which was ultimately negotiated down to 15% with NVIDIA CEO Jensen Huang. This arrangement is widely described by legal scholars and industry experts as "highly unusual" and "unprecedented," with some raising constitutional concerns about it potentially violating the prohibition on taxes on exports. The administration, however, frames the 15% payment as a condition for obtaining export licenses rather than a direct tax.
For NVIDIA (NASDAQ: NVDA), the potential financial implications are substantial. The company had previously estimated a potential loss of an additional $5.5 billion due to tight export controls on its H20 chips. With CEO Jensen Huang estimating $15 billion from H20 chip sales to China, the 15% cut implies a $2.25 billion payment to the U.S. government from NVIDIA alone. Similarly, AMD (NASDAQ: AMD) reported an $800 million charge in Q2 2025 related to restrictions on its MI308 chip sales to China. While the deal allows them to regain market access, their profitability from these specific sales will be significantly reduced.
Initial market reactions have been telling. Despite the re-issuance of export licenses, which would typically be viewed as positive news, shares of both NVIDIA and AMD experienced slight declines in premarket trading after the revenue-sharing deal was reported. NVIDIA's shares dropped 1.16%, and AMD's slipped 2.3% in premarket trading on Monday, an unusual reaction given the granting of export licenses. This unusual reaction underscores investor concern over the precedent set by the revenue-sharing model and its potential long-term impact on future profitability and the broader semiconductor industry's relationship with geopolitical policies.
Winners and Losers in the Unprecedented China Chip Deal
The U.S. government's novel agreement with NVIDIA (NASDAQ: NVDA) and Advanced Micro Devices (NASDAQ: AMD) to allow continued sales of certain AI chips to China, albeit with a 15% revenue share, creates a complex web of winners and losers, distinct from the more direct challenges faced by companies like Micron Technology (NASDAQ: MU).
For NVIDIA (NASDAQ: NVDA) and Advanced Micro Devices (NASDAQ: AMD), the immediate assessment is mixed. On one hand, they are partial, short-term winners. The deal provides a crucial lifeline, allowing them to regain access to the lucrative Chinese market for their "detuned" AI chips—NVIDIA's H20 and AMD's MI308. China represents a significant revenue stream, and this agreement helps them recoup some of the billions in lost sales anticipated due to earlier export restrictions. NVIDIA, for instance, had estimated an $8 billion hit from export restrictions and had already taken a $4.5 billion charge on H20 inventory. This deal, while costly, prevents a complete lockout from a vital market.
However, they are also clear losers in terms of profitability. The 15% revenue share directly impacts their profit margins, effectively increasing their cost of doing business in China. Furthermore, while the U.S. government has granted export licenses, China has simultaneously advised its domestic companies, particularly those involved in government-related work, to avoid using NVIDIA's H20 chips and to favor local alternatives. This counter-pressure from Beijing could significantly limit the actual sales volume and long-term effectiveness of this "deal" for NVIDIA and AMD, especially since their most advanced chips remain prohibited from sale in China.
The U.S. Government emerges as a winner in terms of a new, unprecedented revenue stream from the sales of these chips. This could be framed as a way to benefit the U.S. from continued trade with China while maintaining some level of control over technology exports. However, the deal has drawn criticism from national security, economic, and legal experts who question its constitutionality and worry it sets a dangerous precedent. Critics argue it could undermine the credibility of U.S. export controls, which were initially based on national security concerns, by appearing to prioritize revenue. It also risks alienating allies whom the U.S. has lobbied to impose similar restrictions.
China (Government and Domestic Chipmakers) presents a nuanced picture. In the long term, Chinese domestic chipmakers are potential winners. The ongoing U.S. restrictions and the nature of this "deal" strongly incentivize China to accelerate its efforts towards semiconductor self-sufficiency. By discouraging the use of foreign chips like NVIDIA's H20 and pushing for local alternatives, China aims to reduce its reliance on U.S. technology. This could foster significant growth for domestic Chinese chip companies in the long run, even if it takes time to catch up to global leaders. In the short term, however, China still faces limitations in accessing the most advanced AI chips, which could hinder its progress in cutting-edge AI development. Despite the "deal," the continued reliance on U.S. suppliers for even less advanced chips, coupled with the U.S. government extracting revenue, highlights China's current technological vulnerabilities.
This situation stands in stark contrast to Micron Technology (NASDAQ: MU), a major U.S. memory chip manufacturer, which has faced a more direct and punitive form of retaliation from China. In May 2023, the Chinese government banned the use of Micron's products in critical information infrastructure, citing "serious network security risks." This move was widely seen as a retaliatory measure against U.S. export controls on advanced semiconductors. Micron is a clear loser in this scenario; unlike NVIDIA and AMD, who have found a way to partially re-enter the Chinese market, Micron faced an outright ban in a significant sector. China and Hong Kong accounted for approximately 15% of Micron's total revenue in 2022, and the company anticipated a revenue hit in the low to high single-digit percentage range due to the ban. This direct ban, rather than a negotiated revenue share, highlights China's willingness to target specific companies as a countermeasure in the tech war, especially where it perceives it has alternative suppliers.
A Seismic Shift in Tech Trade: Industry Impact and Broader Implications
The U.S. government's unprecedented agreement with NVIDIA (NASDAQ: NVDA) and Advanced Micro Devices (NASDAQ: AMD), mandating a 15% revenue share from specific AI chip sales in China, represents a seismic shift in global technology trade. This deal, reportedly brokered by former President Donald Trump, allows the companies to resume sales of certain chips (NVIDIA's H20 and AMD's MI308) to China, which were previously subject to export bans, but at a significant cost.
This agreement fits squarely into the broader industry trend of technological decoupling and strategic competition between the U.S. and China, particularly in critical areas like semiconductors and artificial intelligence. It signals a potential new phase in the U.S.-China trade war, where economic concessions are directly tied to market access for sensitive technologies. This "quid pro quo" arrangement is a significant departure from previous export control policies, which typically involved outright bans or licensing without direct revenue sharing. It adds another layer of complexity and unpredictability to global technology supply chains, as companies operating internationally may face increased uncertainty regarding market access and potential revenue-sharing demands in other sectors or regions.
The ripple effects on competitors and partners are substantial. Other semiconductor companies, including Intel (NASDAQ: INTC), which also has a presence in the AI chip market, will be closely watching this development. If this revenue-sharing model becomes a precedent, it could impact their own strategies for selling advanced chips in restricted markets, potentially putting them at a disadvantage if they are not offered similar deals or if the terms are less favorable. While Chinese chipmakers are striving for self-sufficiency, the renewed access to NVIDIA and AMD's chips, even with the revenue share, could temporarily alleviate some pressure on them to develop equivalent high-performance AI chips domestically. However, it also highlights the continued reliance on foreign technology and the imperative for accelerated domestic innovation.
From a regulatory and policy standpoint, the deal has ignited a firestorm of debate. Numerous experts and former government officials have raised serious concerns about the legality and constitutionality of the revenue-sharing arrangement, arguing that it functions as an "export tax," which is explicitly forbidden by the U.S. Constitution. This could lead to legal challenges and set a dangerous precedent for future trade negotiations. This agreement could fundamentally alter the nature of U.S. export controls, moving from a binary "yes/no" licensing system to one that involves financial concessions. This shift could complicate future policy decisions and create a more transactional approach to national security concerns. The mechanism for collecting and utilizing these funds by the Commerce Department, responsible for issuing export licenses, is currently unclear.
Historically, this revenue-sharing agreement is widely described as "unprecedented." U.S. export controls have traditionally focused on restricting the sale of sensitive technologies based on national security concerns, without a direct financial payment from companies to the government for export licenses. While there are no direct historical precedents for this specific revenue-sharing model, the broader context is rooted in decades of U.S. export control policies aimed at preventing the proliferation of technologies that could undermine national security. The recent tightening of AI chip export restrictions, particularly under the Biden administration, aimed to limit China's access to advanced computing power for military modernization and AI development. This new deal, therefore, represents a significant departure from established norms in U.S. trade and national security policy, raising questions about its long-term efficacy and the consistency of U.S. foreign policy.
What Comes Next: Navigating a New Era of Tech Trade
The unprecedented "chip deal" between the U.S. government and semiconductor giants NVIDIA (NASDAQ: NVDA) and Advanced Micro Devices (NASDAQ: AMD), mandating a 15% revenue share from specific AI chip sales in China, ushers in a new and uncertain era for these companies and the broader tech industry. The short-term and long-term implications will shape their strategic pivots, market opportunities, and the challenges they face.
In the short term, the immediate benefit for both NVIDIA (NASDAQ: NVDA) and AMD (NASDAQ: AMD) is the ability to resume sales of their previously restricted chips—NVIDIA's H20 AI accelerator and AMD's MI308—in the lucrative Chinese market. This allows them to recoup a significant portion of lost revenues; NVIDIA had estimated an $8 billion hit due to prior export controls, and AMD reported an $800 million charge related to halted shipments. For instance, if NVIDIA's H20 sales in China return to previous levels, the U.S. government could collect approximately $1 billion per quarter from the 15% share. This re-establishes their market presence and prevents a complete loss of market share to domestic competitors. However, the 15% revenue share will directly impact their profitability from these specific sales, representing a direct hit to their margins. The unprecedented nature of this revenue-sharing agreement has also raised immediate legal and constitutional questions, with some experts arguing it could be construed as an unconstitutional export tax, potentially leading to future challenges.
Looking ahead, the long-term possibilities will necessitate strategic pivots. Both companies have already demonstrated a strategy of designing less powerful, "China-specific" versions of their advanced chips to navigate export controls. This trend is likely to continue, with NVIDIA reportedly working on a version of its new Blackwell series, tentatively named "B20," for the Chinese market, expected in Q2 2025. This requires ongoing R&D investment to balance performance with regulatory compliance. While China remains a critical market, the ongoing geopolitical tensions and regulatory uncertainties may prompt NVIDIA and AMD to further diversify their market focus. AMD, for example, has already seen new opportunities in regions like Saudi Arabia as U.S. export restrictions on advanced tech to China tightened. To mitigate risks and maintain influence, both companies may seek to strengthen partnerships within the global AI ecosystem, including with cloud providers and hardware manufacturers outside of China.
Market opportunities remain significant, primarily driven by China's robust demand for AI chips and heavy investment in AI capabilities. NVIDIA's CEO Jensen Huang has argued that allowing U.S. companies to sell chips in China helps ensure that "America's AI tech stack" remains the global standard, rather than ceding the market entirely to Chinese domestic alternatives. However, significant challenges loom. U.S. export controls have significantly spurred China's domestic semiconductor industry, with companies like Huawei developing their own AI chips (e.g., Ascend series) that are increasingly competitive. This deal might slow, but not halt, China's drive for self-sufficiency. Furthermore, Chinese state media has already begun to cast aspersions on NVIDIA's H20 chips, suggesting they might pose security risks or contain "backdoors," potentially discouraging their adoption by Chinese companies, especially for government-related work. The "deal" is a political arrangement and could be subject to further changes or reversals depending on the evolving U.S.-China relationship and future administrations, creating inherent instability for long-term business planning.
Potential scenarios include this revenue-sharing model setting a precedent for how governments manage the export of sensitive technologies, potentially leading to similar arrangements in other strategic sectors or with other countries. Despite this agreement, the broader trend of technological decoupling between the U.S. and China is likely to continue, as China's long-term goal of achieving self-reliance in semiconductors remains. NVIDIA and AMD will face a more complex competitive landscape, balancing global market leadership with navigating intricate geopolitical regulations and the rise of capable domestic Chinese rivals. The success of this "deal" will ultimately depend on its stability, the market's acceptance of the compliant chips, and the ongoing geopolitical climate.
Conclusion: A New Chapter in the U.S.-China Tech Rivalry
The unprecedented agreement between the U.S. government and semiconductor giants NVIDIA (NASDAQ: NVDA) and Advanced Micro Devices (NASDAQ: AMD), mandating a 15% revenue share from their AI chip sales in China, marks a pivotal and contentious moment in global tech policy. This "chip deal," allowing the resumption of sales for specific AI chips like NVIDIA's H20 and AMD's MI308 to the crucial Chinese market, represents a significant departure from traditional export control mechanisms and sets a new, uncertain precedent.
The core takeaway from this agreement is the introduction of a novel financial arrangement: U.S. chipmakers will remit 15% of their revenue from AI chip sales in China directly to the U.S. government as a prerequisite for obtaining export licenses. This move, widely described as "highly unusual" and "unprecedented," has already sparked concerns about its constitutionality, with some experts likening it to an export tax, which is forbidden by the U.S. Constitution. While NVIDIA (NASDAQ: NVDA) has stated it "follows rules the U.S. government sets for our participation in worldwide markets," the deal fundamentally alters the landscape of U.S. export control.
Moving forward, the market will see a partial restoration of access to the lucrative Chinese market for NVIDIA (NASDAQ: NVDA) and AMD (NASDAQ: AMD), albeit with a direct financial cost. China represents a substantial portion of revenue for both companies, and while the 15% revenue cut will impact profitability, it is viewed by some as a "minor speed bump" compared to a complete sales ban, allowing them to recoup some of the billions in lost sales due to prior restrictions. However, the deal primarily covers less powerful, "obsolete" chips, suggesting that the most advanced AI chips remain off-limits to China. This dynamic is likely to further incentivize China's domestic chip development and reliance on local alternatives, such as Huawei's competing AI chips, potentially reducing its dependence on U.S. suppliers over time.
This agreement carries profound significance for U.S. export control policy and U.S.-China tech relations. It introduces a new, revenue-sharing model for export licenses, potentially setting a controversial precedent for how the U.S. government manages trade with strategic competitors. Critics argue it blurs the lines between national security and economic policy, potentially undermining the credibility of export controls as a national security tool. The constitutional debate surrounding export taxes will likely lead to legal challenges, questioning the durability of this arrangement. Furthermore, the deal could complicate U.S. efforts to rally allies in imposing similar restrictions, as it introduces a unique financial component that other nations may be hesitant to adopt. The move signals a complex and evolving U.S. strategy towards China, balancing national security concerns with the economic interests of American companies.
Investors in NVIDIA (NASDAQ: NVDA), AMD (NASDAQ: AMD), and the broader semiconductor industry should closely monitor several key factors. Firstly, the legal challenges to the deal's constitutionality will be critical, as a successful challenge could invalidate the revenue-sharing component. Secondly, the actual financial impact on NVIDIA and AMD's bottom lines from the 15% revenue share, and how it affects their overall profitability and growth projections, will be important. Thirdly, observe China's response, particularly its acceleration of domestic chip production and potential retaliatory measures, which could further reshape the global semiconductor supply chain. Finally, investors should watch for any indications of whether this "unusual arrangement" becomes a template for future U.S. government interactions with other industries or countries, or if it remains an isolated case driven by the unique dynamics of the U.S.-China chip rivalry. The geopolitical landscape and the ongoing tech competition between the U.S. and China will continue to be a dominant force shaping the semiconductor market for years to come.
