Chinese technology giant Huawei announced the unveiling of over 10 new chipsets designed for AI computing infrastructure on Thursday. The company aims to challenge the market dominance of Nvidia Corp., Advanced Micro Devices (AMD), and Intel Corp. (INTC) with this new lineup.
The new product suite includes next-generation AI accelerators, central processing units (CPUs), and high-speed connectivity chips. Huawei’s rotating chairman, David Wang, stated that the primary objective of these releases is to assist industries in establishing crucial AI infrastructure and to provide an alternative option for China and the world.
Specific product details include the Ascend 960 DT AI accelerator, which is on track for an official launch in the first quarter of 2027. Huawei reports that development is progressing ahead of schedule, with performance expected to be twice that of its predecessor. The Ascend 960PR is slated for release in the third quarter of 2027. During a keynote at the annual Huawei Connect event in Shanghai, Wang highlighted the Ascend chip’s importance to the broader system and confirmed that the company plans to maintain a one-generation-per-year pace for Ascend product development.
Eric Xu, another rotating chairman at Huawei, noted that domestic demand significantly exceeds supply. Consequently, the company is currently limiting its aggressive expansion into international markets. However, Huawei is testing and providing limited supply to several countries with strong demand, Xu said.
The competitive landscape in the AI chip industry is heating up. Chinese AI chipmaker Enflame Technology, backed by Tencent Holdings Ltd. (TCEHY), saw a soaring debut in Shanghai, raising about $912 million in its initial public offering. Enflame is considered one of China’s "four little dragons" of AI chipmaking.
Veteran technology analyst Dan Ives stated that Nvidia remains significantly ahead of Huawei in China. He claimed that even a lower-tier Nvidia chip is roughly one and a half to two years ahead of Huawei’s technology. Ives argued that this technological lead will be difficult for competitors to close as AI demand expands from large-language-model training into areas such as physical AI and autonomous systems.
