Huawei Technologies is making a significant push into the artificial intelligence (AI) chip market with the development of its most powerful processor yet, the Ascend 910D. The company hopes the new chip will rival the offerings of U.S. tech giant Nvidia, specifically aiming to compete with Nvidia’s H100, a leading AI processor used for training complex machine learning models.
The Ascend 910D is expected to outperform its predecessor, the Ascend 910, and take on Nvidia’s high-end chips. Reports from the Wall Street Journal suggest that Huawei has already approached several Chinese tech companies to test the feasibility of this new chip. The company aims to provide an alternative to Nvidia’s dominance in the AI space, particularly in training models—an essential process for machine learning algorithms that rely on massive datasets to improve accuracy.
Huawei’s ambitious move comes amidst ongoing geopolitical tensions, particularly with the United States, which has worked to limit China’s access to cutting-edge AI technology. The U.S. has placed restrictions on China’s acquisition of Nvidia’s most advanced AI chips, including the H100, which was banned from sale in China in 2022. These restrictions have led to a void in the market that Huawei is eager to fill, especially in the wake of its growing ambitions in AI and semiconductor technologies.
According to sources familiar with the matter, Huawei plans to distribute the first batch of samples of the Ascend 910D to select customers as early as late May. Additionally, mass shipments of another advanced AI chip, the Ascend 910C, are expected to commence next month. While the company is aiming to build a robust customer base in China, its success in challenging Nvidia’s position in the AI market remains to be seen.
Despite the restrictions on Nvidia’s products, the U.S. company continues to lead in AI chip technology, particularly in the area of training large language models. Huawei’s efforts to catch up will likely shape the future of AI development in China and could have far-reaching implications for the global tech industry.
