Nvidia, once riding high on a staggering 174% stock surge last year, has encountered headwinds, witnessing its stock price plummet by 19% this year. This downturn has been intensified by the U.S. government’s decision to restrict the export of Nvidia’s H20, a lower-tier artificial intelligence (AI) semiconductor, to China. Further complicating matters, reports have emerged suggesting that Chinese tech giant Huawei is developing AI chips capable of rivaling Nvidia’s offerings. In recent trading, Nvidia’s stock closed at $108.73, reflecting a 2.05% decrease, and an overall year-to-date decline of 19.03%. This marks a significant drop of nearly 27% from its peak of $148.88 on November 7, 2024.
Market Valuation and Recent Setbacks
The company’s market capitalization has consequently contracted to $2.653 trillion, a substantial reduction from last year when it briefly surpassed $3.6 trillion. Nvidia disclosed earlier this month that it had received notification from the U.S. government stipulating that exporting H20 chips to China would require explicit authorization, without a specified termination date for this regulation. The U.S. government’s justification for this measure is rooted in concerns that H20 chips could be exploited in Chinese supercomputers, thereby facilitating advanced AI research.
The U.S. government argues that the aggregation of numerous H20 chips could potentially enable high-performance AI applications. While the H20 chip’s individual computational power might be somewhat limited, its robust connectivity with high-speed memory and other chips makes it suitable for supercomputer construction and utilization.
The H20 Chip: A Response to Export Controls
The H20 chip was, in reality, designed as a lower-performance variant specifically intended for export to China, with the aim of circumventing U.S. export restrictions. Although the H20’s performance is slightly inferior to Nvidia’s state-of-the-art Blackwell AI chip, it can be equipped with the same high-bandwidth memory (HBM), thereby enhancing certain aspects of its overall performance capabilities.
As a result, major tech companies in China, including Tencent and Alibaba, had engaged in a competitive race to secure supplies of H20 and H800 chips. It is worth noting that the Chinese company DeepSice, which triggered an almost 18% single-day stock plunge for Nvidia in January, also employs Nvidia’s lower-tier H20 chips in its operations.
The recent export restrictions imposed on H20 chips are projected to cost Nvidia $5.5 billion in revenue during the first fiscal quarter (February to April). JP Morgan estimates that this measure could potentially wipe out up to $16 billion in Nvidia’s sales this year, accounting for losses related to inventory, unfulfilled purchase agreements, and associated provisions stemming from export limitations.
Reports indicate that Chinese tech giants and AI startups, including Tencent, Alibaba, and ByteDance, have collectively placed orders for over 1.3 million H20 chips, valued at $16 billion. The U.S. government’s export restrictions will effectively block these shipments from being fulfilled. This represents a significant disruption to the existing supply chain and planned AI infrastructure development within China.
A History of Export Regulations
The U.S. government has a long-standing history of imposing regulations on the export of advanced semiconductors to China. Since 2023, Nvidia’s sales in China have declined to less than half of their previous levels as a direct consequence of these AI chip export controls. Nvidia’s “H100,” known as the company’s most advanced general-purpose chip, was prohibited from export to China even before its official launch in 2022. These consistent regulations highlight the strategic importance placed on controlling the flow of advanced technology to China.
While China accounts for approximately 13% of Nvidia’s total sales, some analysts suggest that the actual figure may be higher due to the prevalence of chip smuggling into the country. The recent restrictions on H20 exports are expected to further dampen Nvidia’s growth momentum within the Chinese market and could potentially incentivize further illicit trade activities.
If Nvidia were to effectively withdraw from the Chinese market, domestic semiconductor companies like Huawei would likely seize the opportunity to fill the void in the AI chip market. Industry experts suggest that this could inadvertently transfer the Chinese AI market to Huawei, giving the company a significant competitive advantage and further fueling its growth.
Huawei’s Ambitions and AI Chip Development
Indeed, Huawei is actively exploring opportunities arising from the U.S. restrictions on Nvidia’s exports to China. Just over two weeks after these measures were implemented, news has emerged regarding Huawei’s efforts to capitalize on the situation.
According to a report in The Wall Street Journal, Huawei is engaging with select companies to assess the technical feasibility of its new AI chip, known as the “Ascend 910D.” This chip represents a significant step forward in Huawei’s ambition to become a dominant player in the AI semiconductor space.
The initial samples of this Huawei-designed chip are expected to be completed as early as the end of May. Huawei claims that this chip will offer even greater performance than Nvidia’s current H100, a bold claim that could potentially disrupt the existing market hierarchy. Huawei intends to supply approximately 800,000 AI-specific chips to companies such as state-owned telecommunications firms and ByteDance, the parent company of TikTok, demonstrating its strong ties to the Chinese tech ecosystem.
The Wall Street Journal notes that “Huawei is not focused on making individual chips more powerful but rather on building systems that utilize chips more efficiently.” This suggests a strategic approach that prioritizes system-level optimization over raw processing power, potentially offering a more cost-effective and energy-efficient solution for AI applications.
Huawei previously launched a challenge to Nvidia last year, but it faced significant setbacks. The “Ascend 910C,” which was marketed as having performance similar to Nvidia’s H100 upon its release last year, ultimately exhibited substantial performance disparities. These past failures highlight the significant technological challenges involved in competing with Nvidia’s established expertise.
Even Chinese IT companies have largely avoided Huawei’s products, opting instead for Nvidia’s lower-performance chips. This preference reflects concerns about the reliability, performance, and software ecosystem surrounding Huawei’s AI chip offerings. Huawei’s impending launch of a new chip signifies another attempt to challenge Nvidia’s dominance and establish itself as a credible alternative in the AI semiconductor market.
Shifting Market Dynamics and Nvidia’s Response
Despite concerns about tariffs and export restrictions, investments in AI infrastructure continue to expand globally, driven by the growing demand for AI-powered applications across various industries. However, Nvidia’s sales growth is showing signs of deceleration, indicating a potential shift in market dynamics.
While Nvidia’s sales in the fourth quarter of this year increased by 78% compared to the same period last year, this represents the lowest growth rate in seven quarters. The growth rate, which was 262% in the first quarter of last year, decelerated to 122% in the second quarter, 94% in the third quarter, and 78% in the fourth quarter. This slowing growth rate could be attributed to a combination of factors, including increased competition, export restrictions, and a potential saturation of demand in certain market segments.
In response to these challenges, Nvidia may consider expediting the launch of its “Blackwell” products, its next-generation AI architecture. This accelerated launch could help Nvidia maintain its technological lead and address the growing demand for more powerful and efficient AI chips.
According to Taiwan’s Commercial Times, Nvidia plans to begin production of its “Blackwell Ultra” products in May, accelerating the timeline from the original plan. The publication cited sources in the component industry as saying that “Nvidia has established a policy to advance the production of B300 semiconductors to May.”
The B300 is a graphics processing unit (GPU) semiconductor used in Nvidia’s next-generation high-performance AI semiconductor “Blackwell Ultra” series. While some observers have suggested that the launch of Blackwell Ultra could be delayed due to design flaws, the company is reportedly taking steps to expedite the process, demonstrating its commitment to staying ahead of the competition.
The Commercial Times also noted that “Nvidia’s B200 semiconductors have already entered mass production, so the transition to B300 should proceed smoothly.” Related supply chain companies are also accelerating their preparations, ensuring a seamless and efficient production ramp-up for the Blackwell Ultra series.
Optimistic Forecasts Amidst Uncertainty
Despite the seemingly unfavorable environment surrounding Nvidia, including export restrictions and growing competition, some analysts predict that demand for AI chips will remain robust this year and in the coming years, driven by the continued expansion of AI applications across various industries.
Morgan Stanley stated in a recent report that “despite global macroeconomic uncertainties, demand for AI and inference chips will increase,” and they raised their total sales forecast for Nvidia in 2027 from $230.9 billion to $255.5 billion. This bullish forecast reflects confidence in Nvidia’s long-term growth potential and its ability to capitalize on the expanding AI market.
Dismissing concerns such as H20 export restrictions, Morgan Stanley asserted that “Nvidia will continue to generate performance that overwhelms market trends based on artificial intelligence (AI) semiconductor demand,” effectively dismissing concerns about tariffs and export restrictions. This suggests that Nvidia’s technological superiority and strong market position will allow it to overcome these challenges and continue to thrive.
Furthermore, Moore Insights & Strategy has reaffirmed its “overweight” recommendation and a target stock price of $162, emphatically stating that “Nvidia remains the top semiconductor project, and this has not changed.” This unwavering support highlights the continued confidence in Nvidia’s leadership and its ability to deliver strong returns for investors.
Moore added that “the microeconomic impact on Nvidia is very small, especially considering the strong short-term demand and the significant decline in stock prices.” This suggests that the recent stock price decline may be an overreaction to short-term challenges, and that Nvidia’s long-term fundamentals remain strong.
In summary, Nvidia is currently navigating a complex landscape characterized by export restrictions, intensifying competition from companies like Huawei, and evolving market dynamics. While challenges remain, the company’s strategic responses, including the potential acceleration of its Blackwell product launch, and optimistic forecasts from leading analysts suggest that it is well-positioned to maintain its leading role in the AI semiconductor market and continue to drive innovation in the field of artificial intelligence. The company’s future success will depend on its ability to adapt to changing market conditions, overcome regulatory hurdles, and maintain its technological edge over its competitors.