Key Points:
- Nvidia confirmed its $200 billion CPU market forecast includes China, despite ongoing trade tensions.
- The chipmaker expects its new AI-optimized Vera CPU to generate $20 billion in revenue this year.
- US regulators approved licenses to ship H200 chips to China, but Beijing is blocking imports to support local firms.
- Nvidia reported zero data center revenue from China last quarter as it navigates complex geopolitical standoffs.
Nvidia Chief Executive Officer Jensen Huang confirmed on Saturday that the company’s massive $200 billion market forecast for its central processing units (CPUs) explicitly includes China. The statement, reported first by Reuters, highlights the immense importance of the Chinese market to the US chip giant, even as the company navigates a highly volatile geopolitical standoff between Washington and Beijing. Investors are watching the situation closely to see how much of this multi-billion-dollar market Nvidia can actually access under current trade restrictions.
While the world primarily knows Nvidia for its dominant graphics processing units (GPUs) that power the modern artificial intelligence boom, the company is aggressively expanding into general-purpose computing. In March, Nvidia introduced its first standalone AI-optimized CPU, named Vera. Chief Financial Officer Colette Kress told investors during Wednesday’s first-quarter earnings call that the company expects the new Vera chip to generate a massive $20 billion in revenue this year alone.
This massive push into CPUs represents a critical second front for the chipmaker. The expected $20 billion in Vera chip sales sits completely outside the massive $1 trillion the company has already forecast from its Blackwell and Rubin GPU lines between 2025 and 2027. Early adoption of the new CPU has been highly successful. Nvidia recently delivered its very first batches of the Vera processors to major industry players, including OpenAI, Anthropic, SpaceX, and Oracle, to power their advanced artificial intelligence networks.
The news about the Chinese market forecast arrived right after Nvidia posted record-shattering first-quarter financial results. The Santa Clara-based company reported revenue of $81.6 billion for the quarter ending April 26, representing a spectacular 85% jump from the previous year. Of the $75 billion in total data center revenue, major cloud providers like Microsoft and Amazon accounted for roughly half, at $37 billion. The rest came from a rapidly expanding group of neocloud, industrial, and enterprise accounts that more than tripled in volume year over year.
Despite these incredible global sales, China currently remains a major challenge for the company. Historically, China accounted for at least 20% of Nvidia’s total data-center revenue. However, because of strict export controls and escalating political tensions, that share has effectively dropped to zero. Nvidia reported no China data-center compute revenue in its first quarter and is not assuming any in its current second-quarter outlook of $91 billion, illustrating how severely the trade war has impacted its access to the region.
The complex political situation was on full display last week when Jensen Huang joined US President Donald Trump on a high-stakes diplomatic visit to Beijing. Trump held meetings with Chinese President Xi Jinping on May 14 and May 15 to discuss trade and technology. Although the president agreed last December to allow Nvidia to ship its high-powered H200 AI chips to Chinese customers under special licensing terms, Trump acknowledged that Beijing has yet to approve any purchases.
The US Commerce Department recently granted licenses, clearing the way for up to 750,000 H200 chips to be shipped to 10 Chinese tech giants, including Alibaba, Tencent, and ByteDance. This clearance could potentially unlock up to $26 billion in upside revenue for Nvidia. However, officials in Beijing are holding up the import approvals. The Chinese government is actively trying to protect its own domestic semiconductor sector and boost local champions like Huawei Technologies.
Despite the ongoing deadlock, Huang remains highly optimistic that the massive Chinese market will eventually open up to American chip suppliers. Speaking to reporters after his return to Washington, Huang explained that the Chinese government has to decide how much of its local market they want to protect. He stated that over time, he expects the market to open up. The CEO has previously identified China as a massive $50 billion long-term business opportunity for Nvidia.
Nvidia’s aggressive push into the CPU market also serves as a defensive strategy against its largest US customers. Tech giants like Google, Amazon, and Microsoft are collectively pouring more than $700 billion into AI infrastructure this year. Still, they are also building their own custom silicon to run AI models cheaper and faster. By introducing its own high-efficiency CPUs, Nvidia hopes to dominate the crucial “inference” stage of artificial intelligence, where models serve answers to users in real time.
The coming months will test whether Nvidia can successfully bridge the geopolitical divide and unlock its massive Chinese targets. If the market opens and Beijing approves H200 shipments, the company will secure a massive new revenue stream. Until then, Nvidia must rely on rapid technological innovations, such as the Vera CPU, and strong demand from Western tech giants to keep its historic growth trajectory on track.
