Huawei claims China’s tech now rivals US despite restrictions

Photo:Reuters

China’s technology industry has taken a big step forward in the face of tough challenges from the United States. A senior leader at Huawei Technologies announced that the country has almost completely overcome years of restrictions placed on it by Washington. This achievement, according to Huawei, is proof that Chinese companies can now build and use their own advanced computing systems, artificial intelligence (AI), and software that can match those developed in the US.

Huawei, one of China’s largest and most important technology firms, was placed on America’s trade blacklist back in May 2019. That action was meant to block the company’s access to vital technologies such as advanced chips and software from US suppliers. Many people believed this move would severely weaken Huawei and slow down China’s progress in high-tech fields.

But according to Tao Jingwen, Huawei’s head of quality, business processes, and information technology management, the company has successfully built an entire ecosystem that does not depend on American technology. Tao made this statement during a recent event in Guiyang, a city in China’s Guizhou province. He described the company’s journey as one filled with difficulties but also marked by great resilience and determination.

Stay informed with the latest news. Follow DXB News Network on WhatsApp Channel 

Tao explained that Huawei’s independence is not only a victory for the company but also for China’s broader goal of becoming self-sufficient in technology. He predicted that, with its vast economy and wide range of industries, China will soon surpass the United States in applying artificial intelligence in everyday business and services.

This is a bold claim, but Huawei has backed it up with concrete achievements. On the same day as Tao’s speech, Huawei revealed that users of its cloud platform now have access to one of the most powerful AI systems in the world. Called CloudMatrix 384, this system connects 384 of Huawei’s Ascend AI processors across 12 computing cabinets and four bus cabinets. Together, they provide 300 petaflops of computing power – meaning 300 quadrillion calculations per second – and 48 terabytes of high-speed memory.

For comparison, Huawei said its CloudMatrix 384 can compete with Nvidia’s NVL72 system, which is currently seen as the gold standard in AI computing. This system is designed to handle massive workloads, such as training large AI models, without running into the bottlenecks that usually slow down data centres. Huawei has already deployed CloudMatrix 384 in major data centres across Anhui province, Inner Mongolia, and Guizhou.

Huawei is also pushing ahead in the area of AI models. Its cloud platform now integrates a variety of Chinese-developed AI models, including its own Pangu model and others created by companies such as DeepSeek, Alibaba Cloud, and Moonshot AI. These models are trained to perform different tasks, and Huawei’s platform also offers AI agents that can automatically carry out assignments for human users.

In another major announcement, Huawei introduced a new system called xDeepServe, built specifically for handling large language models (LLMs). Unlike most LLMs that use Google’s Transformer technology, Huawei’s new “Transformerless” design allows the models to be broken down into smaller, more manageable units. The company said this architecture can reduce costs and improve efficiency for businesses that want to use AI on a large scale.

Huawei also displayed three new memory products – OceanDisk EX 560, OceanDisk SP 560, and OceanDisk LC 560 – aimed at making AI computing more efficient. With these products, Huawei is trying to reduce the heavy demand that AI systems place on memory and storage, which is one of the biggest challenges in the industry.

Earlier this month, Huawei took another surprising step by announcing that it would make its Compute Architecture for Neural Networks (CANN) open source. This is a software toolkit that developers can use to build applications for Huawei’s Ascend AI processors. Many observers compared this move to Nvidia’s popular CUDA toolkit, which is widely used across the global AI industry. By opening up its system, Huawei is encouraging more developers in China and abroad to build applications that work on its hardware.

These big developments come at a time when Huawei is also restructuring its cloud division. This division plays a central role in the company’s AI model development, cloud computing, and data centre services. The restructuring is expected to help the company focus more strongly on its long-term goals in AI and cloud technologies.

Tao Jingwen stressed that all of these achievements show how China’s tech industry has managed to survive and grow despite foreign pressure. He suggested that the country’s push for technological independence is not only about survival but also about reaching a point where it can lead the world in certain areas, especially AI.

Experts believe this moment is crucial for the global technology landscape. For years, the US has dominated in areas like advanced chips, cloud computing, and AI. But China’s determination to create its own ecosystem is changing that balance. If Huawei’s claims are accurate, China is no longer just catching up – it is starting to compete head-to-head with American companies.

This development also has geopolitical importance. The US restrictions were originally intended to limit China’s ability to become a world leader in high-tech fields. But instead, those restrictions may have pushed Chinese firms to innovate faster and invest more heavily in building their own alternatives.

Huawei’s progress could inspire other Chinese tech companies to follow the same path. Already, companies like Alibaba Cloud and Moonshot AI are developing their own powerful models and systems. Together, these firms are creating a strong domestic ecosystem that reduces China’s dependence on Western technology.

In short, Huawei’s latest announcements are not just about new products or systems. They represent a broader shift in the global tech race. If China continues on this path, the rivalry between Chinese and American tech companies will only intensify. And as Tao Jingwen confidently suggested, the day may come when China leads the world in applying artificial intelligence to real-world industries and services.