China orders state-funded data centres to use domestic artificial intelligence chips

China has directed state-funded data centre projects to rely exclusively on domestically manufactured artificial intelligence chips and told under-construction sites to remove foreign-made components, a move that targets suppliers such as Nvidia, AMD and Intel.

Chinese regulators have issued guidance requiring all state-funded data centre projects to use domestically manufactured artificial intelligence chips, according to reporting cited in the article. Projects that are less than 30% complete were instructed to remove any installed foreign-made chips or cancel pending orders, while more advanced builds will be evaluated on a case-by-case basis. Reuters is quoted as the source for the directive, which is described as part of Beijing’s accelerated push toward chip self-sufficiency in critical infrastructure.

The policy directly affects major foreign suppliers that sell data centre processors to China, including Nvidia, AMD and Intel. Analysts cited in the article say the guidance could effectively shut Nvidia out of an important market and create opportunities for domestic rivals such as Huawei technologies. The rules explicitly cover Nvidia’s H20 chips, which are the most advanced models currently permitted for sale to China, and reference higher-performance processors such as the B200 and H200, which are barred under existing U.S. export rules but have reportedly circulated through grey-market channels. Nvidia chief executive jensen huang has urged Washington to relax export restrictions, arguing limited access preserves U.S. influence in the global artificial intelligence race.

The move comes amid broader tensions over U.S. export controls aimed at limiting technology transfer that could benefit the Chinese military. Beijing has previously discouraged purchases of top-end foreign chips and showcased data centres running entirely on homegrown artificial intelligence processors. The article notes that many Chinese artificial intelligence data centre projects have received public financing since 2021, suggesting the guidance could have sweeping implications depending on whether it applies nationwide or only in selected provinces. It also recalls prior actions such as the 2023 ban on micron products from critical infrastructure, which prompted the U.S. memory chipmaker to exit the Chinese server market.

72

Impact Score

Saudi Artificial Intelligence startup launches Arabic LLM

Misraj Artificial Intelligence unveiled Kawn, an Arabic large language model, at AWS re:Invent and launched Workforces, a platform for creating and managing Artificial Intelligence agents for enterprises and public institutions.

Introducing Mistral 3: open artificial intelligence models

Mistral 3 is a family of open, multimodal and multilingual Artificial Intelligence models that includes three Ministral edge models and a sparse Mistral Large 3 trained with 41B active and 675B total parameters, released under the Apache 2.0 license.

NVIDIA and Mistral Artificial Intelligence partner to accelerate new family of open models

NVIDIA and Mistral Artificial Intelligence announced a partnership to optimize the Mistral 3 family of open-source multilingual, multimodal models across NVIDIA supercomputing and edge platforms. The collaboration highlights Mistral Large 3, a mixture-of-experts model designed to improve efficiency and accuracy for enterprise artificial intelligence deployments starting Tuesday, Dec. 2.

Contact Us

Got questions? Use the form to contact us.

Contact Form

Clicking next sends a verification code to your email. After verifying, you can enter your message.