Days after the US government ordered federal civilian agencies to strip DeepSeek’s AI technology from their systems over national security concerns, the Chinese AI Lab closed approximately $15 billion in fresh funding, one of the largest financing rounds in Chinese tech history, Bloomberg reported. The round underscores the actual impact US actions may have on DeepSeek’s rise, considering the world continues to look for alternatives to a US-dominated AI space.
The Hangzhou-based lab raised 100 billion yuan from investors including CATL — one of the world’s most dominant battery manufacturer — and Tencent, according to Bloomberg. The round came together quickly and at a larger scale than initially planned: DeepSeek had originally targeted 50 billion yuan, raised that to 80 billion yuan, and ultimately closed at 100 billion. The details have not been publicly confirmed by the company.
This is DeepSeek’s third significant fundraise in under a year. It raised over $7 billion at a $50 billion valuation in June, was reported to be seeking an additional $1.5 billion in July at roughly $71 billion, and has now priced itself into a different category entirely. The company is preparing for an IPO on Chinese markets, likely in 2027. Founder Liang Wenfeng, who previously co-founded the quantitative hedge fund High-Flyer before pivoting to AI research in 2023, retains a staggering 81.6% stake in DeepSeek’s Chinese operating entity.
The fundraise further goes on to show how rapidly DeepSeek is closing the gap between itself and its US counterparts. Anthropic and OpenAI have each raised more than $100 billion. DeepSeek is still well short of that — but at a claimed $15 billion in a single round, from strategically significant investors with deep pockets and clear supply chain relevance, it is no longer a scrappy outlier operating on a fraction of its rivals’ budgets. Even though, that is exactly the story that led to the company’s rapid rise back in early 2025, when its R1 model triggered a $1 trillion selloff in US tech stocks by demonstrating competitive performance at a fraction of the training cost of American models.
DeepSeek is now spending aggressively on data centre infrastructure and broadening its open-source model releases for both domestic and international markets.
The US restrictions add a political dimension to what is already a pointed commercial rivalry. Federal agencies have been ordered to remove DeepSeek’s technology on national security grounds — the same framing applied to TikTok before it, and to Huawei before that. The pattern is consistent even if its ultimate effect on DeepSeek’s international expansion remains unclear. Unlike TikTok, DeepSeek’s models are open-source, which complicates enforcement considerably: removing the app from a government device is straightforward, but open weights distributed across the internet are harder to contain.
What CATL’s participation specifically signals is worth noting separately. The world’s largest battery maker investing in an AI lab is not a passive financial bet — it reflects a broader Chinese industrial strategy of connecting AI capability to physical infrastructure, from EVs to energy storage to robotics. Tencent’s involvement connects DeepSeek’s models to one of the world’s largest consumer and enterprise software distribution networks. Neither investor is simply writing a cheque for returns.