US government urgently processing Nvidia licenses for China but shipping timeline remains unclear

Washington Steps Up Work on Export Licenses

The United States government is working intensively to process export licenses that would allow Nvidia to ship its advanced H200 artificial intelligence chips to China, according to people familiar with the matter cited by Reuters. Officials are reportedly moving “feverishly” through internal procedures after Beijing approved the import of the chips, but the final U.S. green light has not yet been granted.

Under current export control rules, Nvidia must secure a U.S. license before shipping its latest high-performance AI accelerators to Chinese customers, even if those products are specifically designed to comply with American restrictions on computing power and bandwidth.

H200 Chips Cleared by Beijing but Still Waiting on Washington

Chinese regulators have already signed off on Nvidia’s H200 GPUs, paving the way for local companies to purchase the chips once U.S. authorities issue the required export licenses. Despite this regulatory progress in China, Reuters reported that the first shipments are still only expected in 2026, underscoring how geopolitical tension and compliance checks continue to slow the flow of cutting-edge AI hardware across borders.

Nvidia has designed the H200 specifically for markets subject to U.S. export controls, offering data center customers in China a powerful AI chip that stays within Washington’s technical thresholds on performance. The chip is intended for large-scale training and inference workloads in cloud computing, autonomous systems and enterprise AI.

Upfront Payments and Stricter Terms for Chinese Customers

Amid the licensing uncertainty, Nvidia has tightened commercial terms for Chinese buyers of the H200, according to Reuters reporting referenced by industry outlets. The company is now requiring some customers in China to pay the full amount for H200 orders upfront, with no refunds or changes allowed once agreements are signed.

• Full prepayment is being demanded for H200 orders in China.
• Refunds and order changes are reportedly not permitted under the new terms.
• Some buyers may be permitted to use commercial insurance or asset collateral instead of cash.

These tougher conditions mark a shift from Nvidia’s earlier practices, which sometimes allowed partial deposits and more flexible arrangements. They highlight the heightened risk the company faces as it navigates overlapping regulatory regimes in Washington and Beijing.

Balancing Soaring Demand with Geopolitical Risk

Demand for Nvidia’s H200 chips in China is reported to be strong, with local technology companies placing large orders despite the regulatory hurdles and stricter payment rules. Analysts say Chinese cloud providers and AI firms are rushing to secure as much compliant compute capacity as possible to power products such as generative AI models, recommendation systems and industrial automation tools.

For Washington, the licensing process is part of a broader effort to restrict China’s access to the most advanced AI capabilities, particularly for military or security-related applications, while still allowing controlled exports that support commercial activity. For Nvidia, the stakes are high: the company is trying to maintain access to one of the world’s largest AI markets without running afoul of U.S. or Chinese regulators.

These tougher conditions mark a shift from Nvidia’s earlier practices, which sometimes allowed partial deposits and more flexible arrangements. They highlight the heightened risk the company faces as it navigates overlapping regulatory regimes in Washington and Beijing. These tougher conditions mark a shift from Nvidia’s earlier practices, which sometimes allowed partial deposits and more flexible arrangements. They highlight the heightened risk the company faces as it navigates overlapping regulatory regimes in Washington and Beijing.