DeepSeek Excludes Nvidia and AMD from Latest AI Model Access

Feb 26, 2026, 2:45 AM
Image for article DeepSeek Excludes Nvidia and AMD from Latest AI Model Access

Hover over text to view sources

Chinese artificial intelligence lab DeepSeek is making headlines by reportedly withholding access to its upcoming AI model, V4, from major US chip manufacturers Nvidia and AMD. Instead, the lab has chosen to grant early access to domestic suppliers, particularly Huawei Technologies, which has raised eyebrows in the tech community.
The decision to exclude Nvidia and AMD is particularly significant given that DeepSeek's new model is expected to be released around the Lunar New Year. Sources indicate that this exclusion allows Chinese chipmakers a competitive advantage, as they will have several weeks to fine-tune the software for their processors before its global launch.
A senior official from the Trump administration noted that DeepSeek's newest model has been trained using Nvidia's high-performance Blackwell chips. This raises potential concerns about compliance with US export regulations, as the training process reportedly took place on a cluster located in mainland China. The US has imposed strict limits on the export of its most advanced chips to China, including Nvidia's Blackwell chips, which remain restricted.
DeepSeek's strategy may involve obfuscating its reliance on American technology. Reports suggest that the company is considering removing technical indicators that reveal the use of US-made chips in its AI model, opting instead to publicly assert that the training was conducted using Huawei's chips. This tactic could allow DeepSeek to navigate the challenging regulatory landscape while continuing to innovate in AI.
The implications of this move extend beyond just the technology sector. Nvidia, which has reported significant growth in its revenue, remains a key player in the AI chip market. The company recently announced fourth-quarter results that exceeded analysts' expectations, with revenue reaching $68.1 billion, a 73% increase year-over-year. In light of these developments, the exclusion of Nvidia and AMD from DeepSeek's model raises questions about the future dynamics of the AI industry and the ongoing competition between US and Chinese tech firms.
As the landscape of AI technology evolves, the rivalry between domestic and international players is intensifying. The success of DeepSeek's AI models, particularly the upcoming V4, could further catalyze the race for dominance in AI capabilities, underscoring the importance of access to advanced hardware in developing cutting-edge applications.
In summary, DeepSeek's decision to withhold its latest AI model from US chipmakers not only highlights the growing divide in the global tech arena but also poses significant challenges for American firms seeking to maintain their leadership in AI technologies. As the competition heats up, stakeholders in the tech industry will be closely monitoring these developments and their potential ramifications on the future of artificial intelligence.

Related articles

AI Investment Could Transform Jersey's Economy, Experts Claim

Experts suggest that significant investments in artificial intelligence could drive economic growth in Jersey. With proper funding, AI initiatives may lead to job creation and increased productivity, aligning with broader trends in labor demand and workforce transformation.

Ex-Google Engineer Convicted for Stealing AI Secrets to Aid China

Linwei Ding, a former Google engineer, has been found guilty of stealing over 2,000 pages of AI-related trade secrets to benefit Chinese companies. The conviction marks a significant case in the realm of economic espionage, as Ding faces up to 15 years in prison for each count.

Intel's Foundry Talks with Nvidia and Apple Spark Long-Term Industry Concerns

Intel's discussions with Nvidia and Apple regarding future chip manufacturing highlight the company's pivot toward foundry services amidst rising competition. As these talks unfold, questions arise about Intel's capacity to reclaim its position in the semiconductor industry while navigating financial challenges and geopolitical implications.

Former Google Engineer Convicted of Espionage for AI Trade Secrets

Linwei Ding, a former Google engineer, has been convicted of economic espionage and theft of trade secrets related to artificial intelligence technology. The jury's verdict marks a significant case in the ongoing battle against corporate espionage, particularly concerning US technological leadership.

Ex-Google Engineer Convicted for Stealing AI Technology

Linwei Ding, a former Google engineer, was convicted of economic espionage and theft of trade secrets after stealing proprietary AI technology to benefit a startup in China. The conviction highlights ongoing concerns about intellectual property theft linked to foreign interests.