Baidu (09888.HK) shares soared 5.07% in pre-market trading on Friday, following reports that the Chinese tech giant has begun using internally designed chips to train its artificial intelligence (AI) models. This strategic move partially replaces the use of Nvidia's chips and marks a significant shift in China's AI landscape.
According to The Information, citing four people with direct knowledge of the matter, Baidu is experimenting with training new versions of its Ernie AI model using its Kunlun P800 chip. This development comes amid increasing U.S. export restrictions on advanced AI chips to China, prompting Chinese companies to accelerate the development of their own AI chip capabilities.
The market's positive reaction reflects investors' optimism about Baidu's technological advancements and potential cost savings. By reducing reliance on Nvidia's processors, Baidu could strengthen its position in the competitive AI market while mitigating risks associated with supply chain disruptions. However, sources indicate that Baidu has not completely abandoned Nvidia, still using their chips for developing the most cutting-edge models.
China Merchants Securities commented on the broader implications, noting that the rise of AI chip self-development signifies a transition in the AI infrastructure industry from "single GPU supply constraints" to "diversified customized chip solutions." This shift could reshape the competitive landscape and investment logic in the AI sector, potentially benefiting companies like Baidu that are at the forefront of this technological evolution.
Comments