Nvidia exceeded quarterly sales expectations as customers accumulated its AI chips ahead of new U.S. export restrictions to China. However, these restrictions are projected to reduce the company’s sales by $8 billion in the current quarter, leading to a forecast lower than Wall Street estimates on Wednesday.
Despite this, shares of the world’s most valuable semiconductor company rose by 5% in extended trading. Investors were reassured that the impact in the current fiscal second quarter was not as severe as anticipated, and Nvidia emphasized the demand for its new Blackwell chips from clients such as Microsoft. The stock has remained relatively flat this year compared to 2024, when it nearly tripled in value. Nvidia now contends with trade restrictions affecting its sales and a maturing AI data center market.
Washington’s ongoing efforts to limit Beijing’s access to advanced U.S. technology have imposed stricter restrictions on Nvidia’s AI chip exports, limiting access to one of the largest semiconductor markets. During a conference call, CEO Jensen Huang commented on U.S.-China relations, expressing concern about being cut off from China’s significant AI developer base and acknowledging China’s advancing chip industry. He applauded former U.S. President Donald Trump’s decision to rescind a regulation that would have governed global flows of U.S. AI chips.
Huang stated that Nvidia’s Hopper chips could no longer be adjusted for the Chinese market but didn’t mention the Blackwell chips. Reports suggest Nvidia is developing a Blackwell variant for China. Although this may not fully compensate for revenue losses in China, new agreements in the Middle East present growth opportunities, including an extensive data center project in the UAE. Similar deals have also been announced in Saudi Arabia and Taiwan.
Nvidia’s Chief Financial Officer, Colette Kress, noted potential for projects demanding vast amounts of Nvidia AI infrastructure. In the near term, however, export restrictions will be detrimental, as data center revenue in China decreased. The prohibition on selling Nvidia’s H20 chips to China led the company to anticipate a $5.5 billion impact earlier this year, though the actual first-quarter charge was $1 billion less than expected. Nvidia reported a $2.5 billion loss in H20 sales for the first quarter and forecast an $8 billion shortfall in the second quarter.
Nonetheless, the H20 chip generated $4.6 billion in sales during the first quarter, with China contributing 12.5% of total revenue. Analyst Gil Luria from D.A. Davidson remarked that the impact of H20 restrictions was less severe than initially feared.
Major cloud providers, such as Microsoft and Alphabet, have continued their investments in AI data center expansions despite concerns over changing global trade policies. Nvidia reported adjusted earnings of 81 cents per share in the first quarter, with estimates varying widely. Excluding charges, adjusted earnings would have been 96 cents per share.
According to data from LSEG, the adjusted quarterly earnings estimate was 93 cents per share from 17 analysts. Nvidia’s data center segment revenue reached $39.1 billion, slightly below the $39.3 billion estimate.
Nvidia’s manufacturing commitments amounted to $29.8 billion, an increase from the previous year but a decrease from the prior quarter. The company projects second-quarter revenue of $45 billion, plus or minus 2%, compared to an average analyst estimate of $45.90 billion, with projected H20 revenue losses of approximately $8 billion due to export restrictions.
Jacob Bourne, an analyst at Emarketer, noted that trade tensions and potential tariff impacts could pose challenges to AI chip demand. He emphasized that sustaining Nvidia’s dominance requires navigating an increasingly complex geopolitical, competitive, and economic landscape.
