Huawei Ascend 960DT AI chip eyes early 2027 launch

Chinese tech giant Huawei has taken a bold step in the AI ​​”race” by announcing that its next-generation AI chip to rival current Nvidia models will be launched in early 2027. This has the potential to free China from the restrictions the United States has placed on its AI development.

The launch of Huawei’s next-generation computing-comparable Ascend 960DT AI chip has been moved from Q3 2027 to Q1, which may not mean much to many of us, but it’s a big deal.

In late August 2022, the United States blocked China from accessing any new Nvidia technology. Because at that time, all of Nvidia’s chip hardware was manufactured in the Taiwan Semiconductor Manufacturing Company (TSMC) megafactory. While protecting Taiwan’s autonomy was cited as the main reason, in retrospect it appears that it had much more to do with a different geopolitical battle: control of AI.

In January 2025, the launch of China’s DeepSeek model took the world by surprise. It was powerful, but it ran on old Nvidia computer technology. Less than three months later, Nvidia CEO Jensen Huang announced a massive move of chip production to the United States.

Basically, access to advanced chips unlocks the computing power of AI, which is really all you need to know. However, this has increasingly been called a modern-day Cold War between China and the United States, even if both countries are conducting completely different “arms races.”

At this week’s Huawei Connect conference, the company announced that it would bring forward the launch of its new chip to early 2027, which has the potential to reshape China’s artificial intelligence ecosystem. Huawei’s rotating and acting president, David Wang, in his opening speech, revealed that the new chip would surpass the power of previous models.

“We are evolving our Ascend chip series on a one-generation-per-year cycle,” Wang said. “In 2028 and 2029, we will launch the Ascend 970 and 980 chips, respectively. Thanks to the Tau Scaling Law, not only will their computing specifications continue to double, but you can also expect huge improvements across the board in terms of memory bandwidth, memory capacity, interconnect bandwidth, and more.”

He also presented Huawei’s plan to boost the world of AI (or AGI) agent from China, which we haven’t heard much about. You can read about it here, but it essentially underlines how differently the United States is approaching the AI ​​era in comparison. While the focus has been on computing power in the West, with cutting-edge companies like OpenAI and Anthropic leading the way, Chinese engineers have had to design AI infrastructure differently with the chips they have had access to. So in part (not entirely) the ecosystem has prioritized open source innovation.

However, even if 100,000 software engineer brains might be better than the 100 or so currently at Anthropic, there are limitations that come down to the hardware. This was revealed in a leaked call by DeepSeek founder Liang Fengweng with investors in July.

However, the Ascend 960DT chip could change this.

We’ll have to wait and see if it can match Nvidia’s Blackwell GPU chip, whose company founder Jensen Huang said this week is expected to double its sales by 2027.

Interestingly, in his keynote conclusion, Wang noted that while AI “may well be the last technological revolution in human history,” its advancement to benefit societies around the world is based on collaboration, not technological isolationism. “No company alone,” he said, “can build a smart world alone.”

And even in the wake of this week’s news that China has opened the world’s first “smart factory” where robots build robots, people are reportedly much more optimistic about the future of AI and its impact on their lives, compared to the United States. This may be an overlooked aspect of the so-called AI arms race that could shape it as much as the technology itself.

Source: Huawei