KEY FACTS: Huawei, a leading Chinese technology firm, is set to test its new AI chip, the Ascend 910D, in a bid to rival Nvidia’s dominant H100 processor, with testing scheduled to begin in the coming weeks and sample chips expected to reach Chinese tech firms by late May 2025. This move capitalizes on U.S. export restrictions that have limited China’s access to Nvidia’s advanced AI chips, creating an opportunity for Huawei to supply domestic alternatives like the Ascend 910B and 910C, with mass shipments of the latter planned for May 2025. Supported by Beijing’s push for semiconductor self-reliance, Huawei is also developing integrated computing systems like the CloudMatrix 384 to enhance its chips’ performance, signaling China’s broader ambition to challenge U.S. dominance in the AI hardware market amid escalating trade tensions.

---

Source: Huawei, Nvidia

---

Huawei to Challenge Nvidia with New "Ascend 910D" AI Chip

Chinese technology giant Huawei is gearing up to test its latest AI chip, known as the Ascend 910D, which it hopes will rival the high-performance processors of U.S. semiconductor leader Nvidia. This development will challenge U.S. dominance in the artificial intelligence (AI) hardware market. According to a report by The Wall Street Journal on April 27, 2025, Huawei is poised to begin testing this advanced chip in the coming weeks, with the first batch of sample chips expected to reach Chinese tech firms by late May.

The Ascend 910D represents the latest iteration of Huawei’s Ascend AI processor series, a line of chips designed to power AI training and inference tasks. Huawei is betting that this new chip will surpass the capabilities of Nvidia’s H100, a high-end graphics processing unit (GPU) released in 2022 and widely used for training large-scale AI models. The H100 has become a cornerstone of the AI industry, powering data centers and AI applications for companies worldwide. Huawei’s ambition to compete with such a formidable rival underscores the company’s determination to carve out a significant share of the global AI hardware market, particularly within China, where U.S. export restrictions have limited access to Nvidia’s cutting-edge chips.

Huawei has already begun reaching out to domestic technology companies to collaborate on testing the Ascend 910D’s technical feasibility. These tests are critical to ensuring the chip’s performance meets the rigorous demands of AI workloads, such as those required for training large language models or deploying AI inference systems. While the development is still in its early stages, the involvement of Chinese tech firms signals Huawei’s intent to build a robust ecosystem around its AI chips, fostering partnerships that could accelerate adoption and refinement of the technology.

The timing of Huawei’s push into AI chip development is no coincidence. The U.S. government has imposed increasingly stringent export controls on advanced semiconductor technologies, particularly those destined for China. In April 2025, the Trump administration tightened restrictions on Nvidia’s ability to sell its H20 AI chip—a less powerful variant designed to comply with earlier export rules—to Chinese customers. These curbs, combined with earlier bans on Nvidia’s more advanced A100, H100, A800, and H800 chips, have created a significant gap in China’s access to high-performance AI hardware. Nvidia recently reported an estimated $5.5 billion in charges related to its AI chip inventory, largely due to these export restrictions.

For Huawei, these restrictions present a golden opportunity. The company is not only testing the Ascend 910D but also ramping up shipments of its earlier AI chips, the Ascend 910B and 910C, to domestic customers. According to Reuters, Huawei plans to begin mass shipments of the 910C chip as early as May 2025, with some shipments already underway. The 910C is said to offer performance comparable to Nvidia’s H100, making it a viable alternative for Chinese AI developers and companies like ByteDance, the parent company of TikTok, which are seeking domestic solutions to power their AI initiatives.

Huawei’s efforts extend beyond individual chip development. The company is also focusing on creating integrated computing systems to maximize the performance of its AI chips. In April 2025, Huawei unveiled the CloudMatrix 384, a computing system that connects multiple Ascend 910C chips to deliver enhanced performance through parallel processing. Rather than relying solely on making individual processors more powerful, Huawei is leveraging advanced chip arrays to achieve greater efficiency and speed, a strategy that could give it an edge in large-scale AI applications.

Beijing has played a significant role in supporting Huawei’s ambitions. The Chinese government has actively encouraged domestic AI developers to prioritize locally made chips, aligning with President Xi Jinping’s call for “self-reliance and self-strengthening” in AI and semiconductor technologies. This push comes as part of China’s response to U.S. sanctions, which have targeted not only Huawei but also other Chinese firms like SMIC, the country’s leading semiconductor manufacturer. Despite these challenges, Huawei’s progress demonstrates the resilience of China’s semiconductor industry, which continues to innovate under pressure.

The news of Huawei’s Ascend 910D has already sent ripples through the global semiconductor market, particularly affecting Nvidia. On April 28, 2025, Nvidia’s stock fell by just over 2% following reports of Huawei’s plans, with some analysts suggesting that the Chinese company’s advancements could erode Nvidia’s dominance in the AI chip market, at least within China. Nvidia’s shares have been under pressure since the start of the year, with a nearly 20% decline in value since January, exacerbated by earlier reports of Huawei’s 910C chip and the success of Chinese AI startup DeepSeek, which has developed models rivaling those of OpenAI at a lower cost.

Analysts have mixed views on the long-term implications for Nvidia. While some, like Morningstar equity strategist Brian Colello, argue that Huawei’s chips may soon surpass the quality of Nvidia’s older H200 chips, others note that Nvidia’s global dominance remains intact due to its technological lead and widespread adoption outside China. However, the loss of the Chinese market could have lasting consequences, especially as Huawei and other domestic players like SMIC scale up production.

Huawei’s AI chip development is part of China’s quest for technological independence. U.S. export controls, aimed at curbing China’s advances in AI and military technologies, have inadvertently accelerated the country’s efforts to build a self-sufficient semiconductor supply chain. Huawei’s ability to produce advanced chips like the Ascend 910D, reportedly at the 7-nanometer node, has raised questions about the effectiveness of U.S. sanctions. Reports suggest that Huawei has accessed critical foreign tools and sub-components, possibly through intermediaries, to sustain its chip production. This has prompted scrutiny of companies like TSMC, which has acknowledged challenges in ensuring export control compliance among its customers. The resilience of China’s semiconductor industry, despite these restrictions, highlights the complexity of enforcing global supply chain controls.

Success in rivaling Nvidia’s H100 could position Huawei as a leading player in the global AI hardware market, while also bolstering China’s domestic AI ecosystem. However, significant challenges remain, including the need to scale production, optimize performance, and compete with Nvidia’s established software ecosystem, which remains a key advantage for the U.S. company.

---

---

Information Sources:

• The Wall Street Journal
• Yahoo Finance
• Cointelegraph

---

---

If you found the article interesting or helpful, please hit the upvote button and share for visibility to other hive friends to see. More importantly, drop a comment below. Thank you!

This post was created via INLEO. What is INLEO?

INLEO's mission is to build a sustainable creator economy that is centered around digital ownership, tokenization, and communities. It's built on Hive, with linkages to BSC, ETH, and Polygon blockchains. The flagship application, Inleo.io, allows users and creators to engage & share micro and long-form content on the Hive blockchain while earning cryptocurrency rewards.

---

---

Let's Connect

Hive: inleo.io/profile/uyobong/blog

Twitter: https://twitter.com/Uyobong3

Discord: uyobong#5966

---