China has imposed a de facto blockade on exports of critical components needed for NVIDIA’s H200 GPU, effectively halting production of the high-performance AI accelerator in early 2026
The H200, NVIDIA’s flagship data-center GPU succeeding the H100, relies heavily on Asian supply chains—particularly Chinese firms for certain substrates and Taiwan for CoWoS packaging and TSMC’s 4nm wafers. With China controlling over 80% of global rare-earth processing and key chinese hmb supply nodes, the restrictions have choked the flow of essential inputs, forcing NVIDIA to pause H200 output at foundry partners and warn customers of multi-quarter delivery delays.
NVIDIA has not publicly commented on the specific blockade but confirmed in a January 2026 earnings call note that “geopolitical supply-chain dynamics” are impacting availability of certain next-generation products. Analysts estimate the disruption could reduce H200 shipments by 30–50% in Q1–Q2 2026, pushing customers toward alternatives such as AMD’s MI300 series or NVIDIA’s own H100 inventory.
NVIDIA CEO Jensen Huang Criticizes AI Doomsaying as Harmful to Industry and Society 
ByteDance Plans $14 Billion Spend on Nvidia AI Chips in 2026 Amid Surging Demand 
Nvidia to Boost H200 Chip Production Amid Uncertainty Over China Market 
China’s ‘Little Nvidia’ Moore Threads Warns of Overheating Risks After 723% Share Surge 
Amazon adopts Nvidia tech for future AI chips, rolls out upgraded servers to boost cloud-AI power 
Nvidia CEO slams managers resisting AI push, demands full automation of tasks