DeepSeek, the Chinese AI startup that shook global tech markets in early 2025 with its affordable R1 chatbot, is now at the center of a high-stakes waiting game that could signal whether China can build world-class AI without relying on American chips. For weeks, the tech industry has anticipated the release of DeepSeek's next-generation V4 model, which is expected to handle text, images, and video generation simultaneously. Yet despite rumors and speculation, the model remains unreleased, raising questions about technical hurdles, geopolitical constraints, and whether DeepSeek can transition from American-made processors to domestic Chinese alternatives .

What Is DeepSeek V4, and Why Does It Matter?

DeepSeek started as a side project in 2023 within a hedge fund that had access to powerful Nvidia processors. The company exploded onto the global stage in January 2025 when it released R1, a deep-reasoning chatbot that performed comparably to leading US models at a fraction of the cost. That release sent US tech stock prices tumbling and prompted President Donald Trump to call it a "wake-up call" for American companies .

The upcoming V4 model is expected to be multimodal, meaning it can generate text, pictures, and video, making it significantly more capable than previous versions. Stephen Wu, founder of the Carthage Capital fund, predicted that V4 could again shock US tech valuations. "I expect the upcoming DeepSeek V4 release will not just be a software update; it will be a highly capable, open-source model that handles massive context windows at a fraction of the cost," he stated .

DeepSeek's customizable AI tools have already been widely adopted across China and are gaining traction in emerging markets such as Southeast Asia and the Middle East, suggesting strong demand for whatever V4 delivers .

Is DeepSeek Switching to Chinese Chips, and What Does That Mean?

The delay in V4's release may stem from a fundamental technical challenge: transitioning from Nvidia's processors to chips made by China's Huawei. Tech news outlet The Information reported that V4 can run on Huawei's latest chips, citing five people with direct knowledge of large orders placed by tech giants including Alibaba, ByteDance, and Tencent in preparation for the DeepSeek launch .

If confirmed, such a shift would represent a major milestone for China's efforts to reduce dependence on US technology amid strict American export restrictions on advanced AI chips. Wei Sun, principal AI analyst at Counterpoint Research, explained the significance: "It's important to know because at one level, it is a signal of China's AI self-sufficiency trajectory." However, Sun also noted a critical challenge: transitioning from Nvidia to Huawei chips would require "substantial re-engineering" and could "slow development cycles and introduce performance trade-offs, especially for V4, a model expected to be state-of-the-art" .

The US government has imposed export bans on Nvidia's most powerful AI processors to China, citing national security concerns. This restriction creates genuine pressure on Chinese AI companies to develop alternatives, but it also means any transition away from Nvidia hardware is technically complex and time-consuming.

The Smuggling Allegations Complicating the Narrative

Adding another layer of complexity to the V4 story are unconfirmed reports that DeepSeek may have skirted US export restrictions by obtaining thousands of Nvidia's top-end Blackwell chips through third countries and smuggling them into China . Training advanced AI models requires enormous amounts of computing power, far more than what is needed to run the models once they are built.

Nvidia stated it had not seen evidence of such smuggling and told The Information that "such smuggling seems farfetched." DeepSeek and Nvidia did not respond to requests for comment on these allegations . The uncertainty around how V4 was trained underscores the geopolitical tension surrounding AI development in China.

Nvidia

How to Understand the Geopolitical Stakes of DeepSeek V4

• US Export Restrictions: The United States has banned the sale of Nvidia's most advanced AI processors to China on national security grounds, forcing Chinese companies to either develop domestic alternatives or find workarounds.
• China's Self-Sufficiency Goals: If DeepSeek successfully trained V4 entirely on Huawei chips, it would demonstrate that China can build competitive AI models without American hardware, reducing its vulnerability to future US sanctions.
• Technical Performance Trade-offs: Switching from Nvidia to Huawei chips may require significant engineering effort and could result in slower performance or longer development timelines, which may explain the V4 delay.
• Market Implications: A successful V4 release could further disrupt US tech valuations and accelerate the adoption of Chinese AI tools in emerging markets where cost and customization matter more than brand prestige.

Stephen Wu observed that "the ongoing wait for DeepSeek V4 points to friction in scaling advanced models without unrestricted access to top-tier Nvidia hardware" . This friction is real and visible in the delays.

Stephen Wu

What Other Chinese Companies Are Doing

DeepSeek is not alone in pursuing domestic chip alternatives. Another Chinese AI startup, Zhipu, unveiled an image generator in January that it said had been trained entirely on Huawei chips. Alibaba announced this week that it would open a new data center for AI training and inference in southern China, powered by 10,000 of its own chips and operated by China Telecom . These moves suggest a broader industry shift toward reducing dependence on American hardware.

If DeepSeek successfully trains V4 entirely on Huawei silicon, Wu noted, "it signals a material shift in the geopolitical tech landscape." The question is no longer whether China can build AI; it is whether China can build AI without American chips. V4's release, whenever it comes, will provide a concrete answer .