China has mandated that its data centres use more domestic AI chips to reduce reliance on foreign technology. This policy marks a significant step in Beijing’s efforts to cut dependence on the US as export controls tighten.
According to sources familiar with the matter, publicly owned computing hubs across the country must source over 50% of their chips from domestic producers. This move supports the growth of China’s indigenous semiconductor sector.
Mandate Originates in Shanghai
The mandate follows guidelines proposed by the Shanghai municipality in March last year, which aimed to ensure that domestic computing and storage chips make up over 50% of the chips used in the city’s intelligent computing centres by 2025. The guidelines were part of a broader initiative to enhance AI computing resources in Shanghai, China’s financial hub.
Government agencies, including the National Development and Reform Commission (NDRC) and the Shanghai Communications Administration, supported the policy. These agencies, under the Ministry of Industry and Information Technology (MIIT), backed the initiative as part of China’s broader push to strengthen its AI capabilities.
Nationwide Policy Enforcement
The chip quotas initially set in Shanghai have now become mandatory nationwide. Advisers in the data centre industry confirmed that the Shanghai guidelines, initially voluntary, are now being enforced as part of China’s national policy. This move reflects China’s growing commitment to boosting domestic semiconductor production.
The MIIT and NDRC did not immediately respond to requests for comment on the policy.
Conclusion
With the new mandate, China is prioritizing the use of domestic AI chips in its data centres to reduce reliance on US technology. This policy shift highlights the country’s determination to enhance its AI infrastructure and reduce vulnerability to foreign export controls. By 2025, the country aims to significantly strengthen its indigenous semiconductor sector, ensuring greater self-reliance in the technology space.
