Core Insights - Nvidia's CEO Jensen Huang stated that demand for the H200 chip from Chinese customers is "very high," and the company has restarted the H200 supply chain, indicating a cautious approach to China-related transactions amid tense US-China relations [1][3] - The US's AI export control policy requires Nvidia to pay 25% of its sales in China to the US government, imposing strict regulatory constraints on its operations and profitability in the Chinese market [3] - China's increasing self-sufficiency in mature process chips and growing competitiveness in specific markets like AI chips suggest that the Chinese market is no longer a necessity for US chip companies [4][6] Company and Industry Analysis - The H200 chip features significant performance improvements over its predecessor, the H100, with 141GB HBM3e high-speed memory and a memory bandwidth of 4.8TB/s, making it attractive for Chinese enterprises seeking to enhance AI computing capabilities [6] - The ongoing competition in the tech sector between the US and China has entered a new phase, with China accelerating its efforts to build a self-sufficient chip industry, potentially leading to a fundamental shift in the global high-end chip market [6][8] - Huang's remarks reflect not only commercial insights regarding the H200's transactions but also the complexities and future directions of the US-China tech rivalry, highlighting the need for all parties to prepare for unknown risks in a rapidly changing economic and policy environment [8]
黄仁勋心知肚明,中国只留最后机会,特朗普再搅局,这摊子都得砸