US chipmaker NVIDIA is set to start mass production of a powerful AI chip designed specifically for the Chinese market, according to sources cited by Reuters. The H20 chip, which was developed in response to US restrictions on semiconductor equipment exports to China, will enter production in the second quarter of 2024. Originally slated for launch in November 2023, the chip’s release was delayed due to integration issues. However, Chinese companies have been hesitant to purchase the H20 chip and have been exploring domestic alternatives instead. Chinese search engine giant Baidu, for example, signed a partnership with Huawei to acquire AI chips. This development highlights the challenges faced by US chipmakers in navigating export controls while maintaining their presence in the Chinese market.
The news of NVIDIA’s production plans comes as the value of semiconductor deals in China reached its peak in 2020, as reported by GlobalData. In 2020, Chinese semiconductor deals amounted to $32.9 billion, a significant increase compared to the $3.1 billion in deals recorded in 2019. However, the value of deals decreased to $16 billion in 2021 and further dropped to $14.8 billion in 2023. These figures indicate the fluctuating nature of the Chinese semiconductor market, which is influenced by various factors such as geopolitical tensions and trade restrictions.
The global AI market is projected to see substantial growth, with a value of $383.3 billion expected by 2030. This represents a compound annual growth rate of 21% between 2022 and 2030. The demand for AI technologies and chips is driven by advancements in artificial intelligence applications across industries, including healthcare, finance, and transportation.
As companies like NVIDIA navigate the complex landscape of export controls and market dynamics, their ability to adapt and cater to specific markets will be crucial for their continued success and growth in the global semiconductor industry.
The source of the article is from the blog japan-pc.jp