Huawei has set a goal of competing with Nvidia in artificial intelligence infrastructure, turning to its own chips and computing systems to try to offset China's disadvantage against the United States in processing capacity.
Guo Ping, chairman of Huawei's supervisory board, told new employees at a company meeting that building globally competitive AI agents represents an important opportunity for the firm.
Guo acknowledged that China trails the United States in processing capacity, but argued the country holds advantages in data application and industrial scenarios, and has a talent pool roughly on par with that of the United States, according to the Chinese economic outlet Yicai, which reported the remarks on Wednesday.
Ascend chips and SuperPoD clusters
Huawei is seeking to close that gap through its Ascend processors and an architecture that allows large numbers of chips to be grouped together in systems such as SuperPoD, rather than competing purely on the processing power of individual semiconductors.
Guo specifically pointed to the Ascend 950 and said the company intends to gain an edge through architectural innovations and large computing clusters, aimed at solving practical problems and improving the customer experience.
Huawei's strategy does not involve developing its own AI model. Instead, the company plans to provide the infrastructure needed to run models from different developers through its Ascend, Kunpeng and SuperPoD systems.
Guo said Huawei has no plans to expand its business areas and intends instead to offer diversified options for AI development.
US restrictions on advanced chips
The push carries particular weight given restrictions imposed by the United States on China's access to advanced chips, which have pushed Beijing to accelerate development of domestic alternatives.
Huawei's Ascend processors have become one of the main Chinese rivals to Nvidia's processors.
Wider debate over AI development
Guo's comments come amid an international debate over AI, after figures including Dario Amodei of Anthropic, Sam Altman of OpenAI and Elon Musk of xAI recently called for the technology's development to slow down for safety reasons. US President Donald Trump rejected that possibility, arguing it could give China an advantage in the technology race.
The debate has again placed the competition between Beijing and Washington at the center of discussion over the future of AI, both in terms of technological leadership and the possibility of establishing shared mechanisms to control the risks associated with the technology.
