The relaxation of the United States\u2019 export controls for Nvidia\u2019s H20 chips won brief applause in China \u2013 while raising longer-term concerns about whether Chinese firms will over-rely on foreign artificial intelligence chips.<\/p>\n
During a trip to China on July 15, Nvidia Chief Executive Jensen Huang said<\/a> in a press conference that the company will resume H20 chip sales to China, now that the US government has indicated it will soon grant export licenses.<\/p>\n \u201cI hope to get more advanced chips into China than the H20,\u201d Huang said. \u201cTechnology is always moving on. Today, Hopper\u2019s terrific, but some years from now, we will have more and more and better and better technology, and I think it\u2019s sensible that whatever we\u2019re allowed to sell in China will continue to get better and better over time as well,\u201d he said.<\/p>\n He also took the opportunity to praise Huawei\u2019s achievements in making AI chips.<\/p>\n However, some Chinese commentators viewed the development as unfavorable for China\u2019s chip-making sector.<\/p>\n \u201cThis is not a simple lifting of the export restrictions, but a carefully designed measure for the United States to maintain its technological blockade against China,\u201d a Guangdong-based columnist says in an article.<\/p>\n \u201cThe H20\u2019s FP16 computing power is only 15% of H100, while its NVLink bandwidth is reduced from 900GB\/s to 400GB\/s. The chip\u2019s transformer engine (TE) is completely deleted,\u201d he says. \u201cSuch a design ensures the chip\u2019s AI inference ability and reduces its AI training capability, perfectly implementing the United States\u2019 strategy of blocking high-end chips, but not mid-end ones, to China.\u201d\u00a0\u00a0<\/p>\n \u201cBy limiting the key performance of H20, the US can maintain its blockade of high-end computing power while handing Chinese companies a glass of \u2018poisoned wine,\u2019\u201d he says.<\/p>\n In Chinese idiom, a person who \u201cdrinks poisonous liquor to quench thirst\u201d knows that it will kill him, in the long run, but he can\u2019t do anything to change the situation. Applied to Chinese technology companies, this means that the domestic chipmakers can benefit from foreign AI chips in the short term but will miss an opportunity to grow and establish an ecosystem.<\/p>\n An AI firm needs tens of thousands of Nvidia\u2019s high-end chips, such as A100 or H100, to train a large language model (LLM) like ChatGPT. Once an LLM is developed, the company can use slower graphics processing units for inference tasks.<\/p>\n In an article<\/a> published by\u00a036kr.com, a columnist using the pseudonym \u201cSilicon Rabbit\u201d says that Nvidia\u2019s Huang made a subtle and cunning move to help the US curb China\u2019s chip sector.\u00a0<\/p>\n \u201cImagine that someone sold you a Ferrari with a powerful V12 engine, but downgraded its gas pipe, gearbox and wheels. This car can run normally on straight roads, but it faces limitations when continuously speeding up or making sharp turns,\u201d he says, attributing the metaphor to an unnamed senior software engineer who had participated in the Hopper architecture\u2019s performance optimization project.<\/p>\n He says that the computing power of a single H20 chip is far below that of the H100, while a reduced interconnect bandwidth means a significant reduction in AI training capability.<\/p>\n \u201cAI training is similar to having tens of thousands of people work together, which requires fast information exchanges,\u201d he says. \u201cA low interconnect bandwidth means that people communicate slowly, resulting in a low thinking efficiency.\u201d<\/p>\n He says that the H20 chip cannot be used to train trillion-parameter LLMs.\u00a0<\/p>\n \u201cThe H20 generously offers 96 gigabytes of the third-generation high bandwidth memory (HBM3) \u2013 higher than the H100\u2019s 80GB HBM3e. However, the H20\u2019s memory bandwidth is only 4.0 terabytes per second (TB\/s), lower than the H100\u2019s 4.8 TB\/s,\u201d he says. \u201cIt is like someone giving you a bigger table to read more books, but making it harder for you to take books from the shelves.\u201d<\/p>\n The writer says the relaxation of the export rules for the\u00a0H20 is aimed at permitting the\u00a0US to control the pace of China\u2019s AI development precisely.<\/p>\n Huawei\u2019s 910B<\/p>\n Reuters, citing sources familiar with the situation, reported<\/a> that Chinese internet giants, including ByteDance and Tencent, are submitting applications for the H20 chip. ByteDance denied the report. Tencent did not respond to Reuters\u2019 request for comment.<\/p>\n On April 9, 2025, according to Nvidia, the US government informed the company that a license is required for exporting its H20 products into the Chinese market. The new curb was a part of Washington\u2019s countermeasures after China retaliated against the Trump administration\u2019s reciprocal tariffs.<\/p>\n In announcing results for the three months ended April 27, Nvidia said<\/a> sales of H20 products were US$4.6 billion before the new export licensing requirements took effect. It said it could not ship an additional $2.5 billion of H20 revenue.<\/p>\n After US and Chinese officials held meetings in London on June 9, both sides agreed to de-escalate the trade war. Beijing decided to ease export controls on niche metals to the US. In return, the US would allow Chinese firms to use its chip-making software and export parts for China\u2019s C919 flight engines. And now, the US will enable Nvidia to ship the H20 chips to China.<\/p>\n A Guangdong-based columnist says<\/a> Nvidia\u2019s H20 chips will enjoy an advantage in the Chinese market, although Huawei\u2019s Ascend 910B chips perform better in many aspects of AI training. He says Nvidia\u2019s CUDA platform is more advanced than Huawei\u2019s MindSpore framework, making customers reluctant to use non-Nvidia chips.<\/p>\n For example, he says that Alibaba prefers to use the H20 chips to migrate its existing AI system, while the Ascend 910B chips may target state-owned enterprises.<\/p>\n A Beijing-based writer expects<\/a> Nvidia\u2019s CUDA platform to continue enjoying an 80% market share in China, as it would be expensive for companies to switch to new platforms.<\/p>\n On July 18, a spokesperson for the Chinese Ministry of Commerce said<\/a> the US should abandon its \u201czero-sum mentality\u201d and further remove a series of trade restrictions targeting Chinese companies that the ministry considered unreasonable.<\/p>\n The spokesperson stated that in May, the US unveiled export control measures targeting Huawei\u2019s Ascend chips, tightened restrictions on Chinese chip products following unfounded accusations, and intervened in fair market competition with administrative measures. The spokesperson urged the US to work with China to correct erroneous practices through equal consultation.<\/p>\n Meanwhile, the Trump administration increased<\/a> its efforts to prevent China from obtaining Nvidia\u2019s high-end chips. It urged Malaysia and Thailand to curb transhipments of Nvidia\u2019s AI chips to China. <\/p>\n On July 14, the Malaysian government announced<\/a> that export, transshipment or transit of high-performance AI chips of US origin will require a trade permit. Companies must notify the government at least 30 days before shipping Nvidia\u2019s high-end chips elsewhere.<\/p>\n