🤖 AI Summary
Nvidia CEO Jensen Huang warned that blunt export controls on AI hardware risk hurting U.S. interests as much as — or more than — China’s. In an interview he said Nvidia has gone “from 95% market share to 0%” in China after U.S. restrictions on advanced AI chips, arguing that half the world’s AI researchers are in China and it’s a mistake to prevent them from building on American technology. Nvidia designed a processor to comply with new limits, but Chinese regulators have reportedly told local firms not to buy U.S.-compliant chips. The company also noted past rounds of U.S. policy: 2022 rules limiting exports of its top accelerators, later licensing decisions and revenue-sharing arrangements for some sales, and reciprocal Chinese limits on rare-earth exports.
The significance is both geopolitical and technical: limiting access to high-performance accelerators reshapes global research dynamics, supply chains, and market share while not eliminating demand. Practically, export caps, licensing regimes and retaliatory resource controls (like rare earths) force chipmakers to redesign products to meet policy constraints, reduce market reach, and alter deployment timelines for large-scale models. Huang said Nvidia now assumes China will be excluded from its forecasts, underscoring how regulation can swiftly reconfigure hardware availability, international collaboration, and competitive leadership in AI.
Loading comments...
login to comment
loading comments...
no comments yet