Nvidia expects an $8 billion revenue hit in the second quarter as U.S. export restrictions choke off its AI chip business in China.
The company confirmed a $4.5 billion charge in the first quarter of its 2026 fiscal year, along with $2.5 billion in unsold H20 chips due to licensing hurdles imposed by the U.S. government. Together, that’s a $7 billion dent in just one quarter.
The restrictions, enforced under the Trump administration and still holding strong, target high-performance AI chips like Nvidia’s H20, a product critical to data centres in China.
CEO Jensen Huang said: “The $50 billion China market is effectively closed to us. The H20 export ban ended our Hopper data centre business in China. We cannot reduce Hopper further to comply.”
For a company riding high on global AI demand, the setback is huge. The second quarter’s projected revenue stands at $45 billion, but Nvidia has already priced in the impact of the licensing blockade.
Nvidia is now pushing a lower-powered GPU dubbed “B20” to meet Chinese demand while skirting around restrictions. Meanwhile, AMD is responding with its own China-targeted chip, the Radeon AI PRO R9700, as firms scramble to protect market share without breaching U.S. export policy.
And yet, these workarounds might not be enough to close the gap. There were recent reports that Nvidia is also developing a budget AI chip based on its latest Blackwell architecture. Estimated to cost between $6,500 and $8,000, it’s still priced well below the flagship H20 GPU, which typically sells for $10,000 to $12,000.
The geopolitical challenge over AI supremacy is changing strategy. Huang stressed that “China is one of the world’s largest AI markets and a springboard to global success with half of the world’s AI researchers based there; the platform that wins China is positioned to lead globally today.”
Huang also welcomed the Biden administration’s decision to abandon a proposed Artificial Intelligence Diffusion Rule, which would have introduced even tighter limits.
But he was clear about the bigger problem: “The question is not whether China will have AI; it already does. The question is whether one of the world’s largest AI markets will run on American platforms. Shielding Chinese chip makers from U.S. competition only strengthens them abroad and weakens America’s position.”
What we’re seeing is a fracture in the global AI value chain. Nvidia, which had become a pillar in this ecosystem, now finds one of its most important markets functionally out of reach. That’s a change no patchwork of scaled-down GPUs can fix.
