Huawei Pulls Its Next AI Chip Forward in a Direct Nvidia Challenge
Huawei Pulls Its Next AI Chip Forward in a Direct Nvidia Challenge
Huawei has brought forward the launch of its next-generation Ascend 960DT AI chip to the first quarter of 2027, from a previously planned third-quarter debut, sharpening its challenge to Nvidia as the AI hardware race intensifies.
The accelerated timetable was announced at Huawei Connect, where the company paired its chip roadmap with a broader pitch: Chinese-made accelerators, linked at vast scale, can become a single AI computing engine. A Huawei spokesperson said the Ascend 960 chips were arriving ahead of schedule, “doubling performance and advancing year by year.”
Huawei’s strategy rests on its Peerium Computing Architecture and UnifiedBus interconnect technology, designed to bind processors, memory, storage and networking into systems for AI training and inference. The company says its Atlas 950 SuperCluster can connect as many as 256,000 accelerator cards. Its newly unveiled chip technologies underline what another report cast as a narrowing AI gap between China and the United States—and an increasingly direct contest with global leaders including Nvidia.
But the announcement also invited scrutiny. China technology analyst Rui Ma noted that Huawei had once described an Atlas 960 SuperPoD scaling to 15,488 Ascend 960 chips, while the latest presentation referenced 4,096. “The chip itself is coming WAY earlier, but the SuperPoD they announced is much smaller than what they originally laid out,” she wrote.
That tension captures Huawei’s position: faster execution, but unresolved questions over system-scale delivery. The company is pressing on despite U.S. limits on China’s access to advanced semiconductor technology. Ma argued those curbs would not halt the push for domestic capacity because “the stakes for self-sufficiency are just too high at this point.” With a Trump-Xi meeting due days after the announcement, Huawei’s revised schedule also lands squarely in the geopolitical contest over who sets AI’s next pace.
Write a comment