DeepSeek raises $15bn from Tencent and CATL ahead of IPO
The Chinese AI lab secured a massive funding round to compete with US giants while facing new federal restrictions in the United States.
The short version
- DeepSeek reportedly raised about 100bn yuan ($15bn) in one of China's largest private funding rounds.
- Major investors in the deal include battery maker CATL and the parent company of WeChat, Tencent.
- The funding round size increased during negotiations, moving from an initial 50bn yuan target to 100bn yuan.
- The US Department of Homeland Security recently ordered federal agencies to stop using DeepSeek products.
- DeepSeek's cloud services currently run on chips manufactured by China's Huawei.
Why it matters
This funding shows how Chinese firms are consolidating resources to challenge US dominance in the AI sector. It also highlights the growing tension between global AI development and national security restrictions.
A massive capital injection
DeepSeek, a prominent Chinese AI laboratory based in Hangzhou, has reportedly secured approximately 100bn yuan ($15bn) in a private funding round. As reported by The Next Web, this deal stands as one of the largest private funding events ever recorded in China. The investment signals a significant move to scale the company's infrastructure and capabilities.
The funding round saw a notable increase in size during the negotiation phase. While the company initially aimed to raise 50bn yuan, that target was later adjusted to 80bn yuan before finally closing at the 100bn yuan mark. This upward trend was driven by high demand from various investors who were eager to participate in the deal.
Key players in this funding round include the battery manufacturer CATL and Tencent, the company that owns the popular messaging app WeChat. This deal represents the second outside funding round for DeepSeek this year, following a previous round in June where the company raised more than $7bn. The lab is also reportedly preparing for a potential stock market listing in the future.
Competing with US giants
The primary goal for DeepSeek is to use this capital to compete more effectively against leading United States AI laboratories. The company gained global attention in early 2025 by demonstrating that high-performing AI models could be developed at a much lower cost than previously thought. This efficiency sparked a sell-off in various technology stocks across the globe.
Despite the massive $15bn raise, DeepSeek still faces a significant funding gap compared to its American rivals. According to Bloomberg, both OpenAI and Anthropic have raised more than $100bn each. This means that while DeepSeek is well-funded, it still has considerably less capital to spend on research and development than the top US firms.
DeepSeek has already begun allocating resources toward expanding its data centers and developing open-source models. These are models that the public can download and adapt for their own specific needs. By focusing on open-source options, the company aims to establish a foothold in the global AI ecosystem while scaling its internal capabilities.
Hardware and infrastructure
The technical backbone of DeepSeek's operations is currently navigating complex international trade restrictions. The company's cloud services already run on hardware chips manufactured by Huawei, which is a Chinese technology firm. This setup is notable because it operates despite current United States limits on the export of advanced chips to China.
The ability to build and run powerful models on domestic hardware is a key component of DeepSeek's strategy. By utilizing Huawei chips, the company can maintain its development pace without relying on restricted Western components. This move underscores the growing independence of the Chinese AI supply chain from American technology.
The lab's founder, Liang Wenfeng, played a central role in establishing the company's early foundation. He also co-founded the Chinese hedge fund High-Flyer, which provided the initial funding for much of DeepSeek's early work. This transition from private hedge fund backing to massive corporate investment marks a new chapter for the company's growth.
National security concerns
The funding round closed just two days after a significant move by the United States government regarding DeepSeek's products. The US Department of Homeland Security ordered federal agencies to stop purchasing any DeepSeek products. Furthermore, the order required these agencies to remove the company's software from their existing systems.
The government's decision was based on specific national security concerns. Officials expressed worry that Chinese authorities could potentially gain access to sensitive United States government data through the use of DeepSeek's tools. This move highlights the increasing friction between international AI development and national security protocols.
While the funding from Tencent and CATL provides DeepSeek with the resources to expand, the US government's actions create a significant hurdle for its growth in the Western market. The company now faces a landscape where its technology is being actively restricted by some of the world's most powerful government entities.
What to watch
- Whether DeepSeek successfully launches its IPO despite US restrictions.
- How other Chinese tech firms respond to the $15bn funding milestone.
- If US chip export limits will be further tightened in response to DeepSeek's hardware choices.
Try it
Look up the current status of federal agencies' software requirements to see how many AI tools are being phased out for security reasons.
Sources and links
- Governmentordered federal agencies federalregister.gov
- ReportingThe Next Web: the article we sourced
Liked it? Say why
Add a line for other readers.
No account needed. We never show your email. Comments with links, ads or abuse are held and not shown. Your comment shows after a quick check.
Listen to the AI Wire headlines
Listen to the AI WireListen to our daily AI shows every morning: on your commute, in the shower, wherever you are
Listen to the daily showsFirst reported by The Next Web · Read the original. Written by the SI Agent News Desk from the reporting above.Written with AI from the source above. Every number is checked against the source. Spot a mistake? Corrections.
