State-linked Chinese entities are using cloud services provided by Amazon.com Inc or its rivals to access advanced US chips and artificial intelligence (AI) capabilities that they cannot acquire otherwise, public tender documents showed.
The US government has restricted exports of high-end AI chips to China over the past two years, citing the need to limit the capabilities of the Chinese military.
However, providing access to such chips or advanced AI models through the cloud is not a contravention of US regulations, as only exports or transfers of a commodity, software or technology are regulated.
Photo: Reuters
A Reuters review of more than 50 tender documents posted over the past year on publicly available Chinese databases showed that at least 11 Chinese entities have sought access to restricted US technologies or cloud services.
Among those, four explicitly named Amazon Web Services (AWS) as a cloud service provider, although they accessed the services through Chinese intermediary companies rather than from AWS directly.
The tender documents show the breadth of strategies Chinese entities are employing to secure advanced computing power and access to generative AI models. They also underscore how US companies are capitalizing on China’s growing demand for computing power.
“AWS complies with all applicable US laws, including trade laws, regarding the provision of AWS services inside and outside of China,” a spokesperson for Amazon’s cloud business said.
AWS controls nearly one-third of the global cloud infrastructure market, according to research firm Canalys.
In China, AWS is the sixth-largest cloud service provider, according to research firm IDC.
Shenzhen University spent 200,000 yuan (US$28,027) on an AWS account to gain access to cloud servers powered by Nvidia A100 and H100 chips for an unspecified project, according to a March tender document.
It got the service via an intermediary, Yunda Technology Ltd Co (成都運達科技), the document showed.
Exports to China of the two Nvidia chips, which are used to power large language models (LLM) such as OpenAI’s ChatGPT, are banned by the US.
Shenzhen University and Yunda Technology did not respond to requests for comment.
Nvidia declined to comment on Shenzhen University’s spending or on any of the other Chinese entities’ deals.
Zhejiang Lab (之江實驗室), a research institute developing its own LLM, GeoGPT, said in a tender document in April that it intended to spend 184,000 yuan to purchase AWS cloud computing services, as its AI model could not get enough computing power from homegrown Alibaba (阿里巴巴).
A spokesperson for Zhejiang Lab said that it did not follow through with the purchase, but did not respond to questions about the reasoning behind the decision or how it met its LLM computing power requirements.
Alibaba’s cloud unit, Aliyun (阿里雲), did not respond to a request for comment.
The US government is trying to tighten regulations to restrict access through the cloud.
“This loophole has been a concern of mine for years, and we are long overdue to address it,” US House of Representatives Foreign Affairs Committee Chairman Michael McCaul said in a statement, referring to the remote access of advanced US computing through the cloud by foreign entities.
ENDEAVOR MANTA: The ship is programmed to automatically return to its designated home port and would self-destruct if seized by another party The Endeavor Manta, Taiwan’s first military-specification uncrewed surface vehicle (USV) tailor-made to operate in the Taiwan Strait in a bid to bolster the nation’s asymmetric combat capabilities made its first appearance at Kaohsiung’s Singda Harbor yesterday. Taking inspiration from Ukraine’s navy, which is using USVs to force Russia’s Black Sea fleet to take shelter within its own ports, CSBC Taiwan (台灣國際造船) established a research and development unit on USVs last year, CSBC chairman Huang Cheng-hung (黃正弘) said. With the exception of the satellite guidance system and the outboard motors — which were purchased from foreign companies that were not affiliated with Chinese-funded
PERMIT REVOKED: The influencer at a news conference said the National Immigration Agency was infringing on human rights and persecuting Chinese spouses Chinese influencer “Yaya in Taiwan” (亞亞在台灣) yesterday evening voluntarily left Taiwan, despite saying yesterday morning that she had “no intention” of leaving after her residence permit was revoked over her comments on Taiwan being “unified” with China by military force. The Ministry of the Interior yesterday had said that it could forcibly deport the influencer at midnight, but was considering taking a more flexible approach and beginning procedures this morning. The influencer, whose given name is Liu Zhenya (劉振亞), departed on a 8:45pm flight from Taipei International Airport (Songshan airport) to Fuzhou, China. Liu held a news conference at the airport at 7pm,
Authorities yesterday elaborated on the rules governing Employment Gold Cards after a US cardholder was barred from entering Taiwan for six years after working without a permit during a 2023 visit. American YouTuber LeLe Farley was barred after already being approved for an Employment Gold Card, he said in a video published on his channel on Saturday. Farley, who has more than 420,000 subscribers on his YouTube channel, was approved for his Gold Card last month, but was told at a check-in counter at the Los Angeles International Airport that he could not enter Taiwan. That was because he previously participated in two
Taiwan was ranked the fourth-safest country in the world with a score of 82.9, trailing only Andorra, the United Arab Emirates and Qatar in Numbeo’s Safety Index by Country report. Taiwan’s score improved by 0.1 points compared with last year’s mid-year report, which had Taiwan fourth with a score of 82.8. However, both scores were lower than in last year’s first review, when Taiwan scored 83.3, and are a long way from when Taiwan was named the second-safest country in the world in 2021, scoring 84.8. Taiwan ranked higher than Singapore in ninth with a score of 77.4 and Japan in 10th with
RSS