DeepSeek has informed prospective investors in its second fundraising round that it is suspending the deal for the time being, according to people familiar with the matter. The decision comes days after comments widely attributed to founder Liang Wenfeng regarding U.S.–Chinese AI competition circulated widely online.

The Chinese AI pioneer verbally notified some would-be backers that investment agreements would not be signed in the coming days as previously expected, the people said. DeepSeek may resume the deal process at a later date, they added, requesting anonymity to discuss private matters.

The suspension stemmed in part from Liang's frustration over online reports concerning his remarks to investors during the company's first financing round, which closed in June and raised $7 billion for the lab. The people noted that Bloomberg has not independently verified the authenticity of those posts, which centered on a transcript of a meeting between Liang and unidentified parties.

Chinese media outlet Yicai reported this week that the billionaire founder discussed a reliance on Nvidia Corp. chips for AI development and China's persistent lag in AI sophistication relative to the United States. The reference to Nvidia hardware underscores the pressure Chinese AI labs face under U.S. export controls, which since 2022 have restricted Beijing's access to the chipmaker's most advanced data center GPUs. DeepSeek did not immediately respond to an emailed request for comment regarding the transcript or the fundraising situation.

Negotiations remain fluid, and the company could still proceed with the deal. It is unclear whether DeepSeek has communicated its intentions to all prospective investors in the current round. The talks began just weeks after DeepSeek closed its record-breaking first financing round, securing major backers including Tencent Holdings Ltd. and Contemporary Amperex Technology Co. Ltd.

DeepSeek was aiming to raise at least 10 billion yuan in additional funds through the follow-on deal, though the final figure could increase depending on the number of participating investors. The startup had been targeting a pre-money valuation of at least 480 billion yuan, Bloomberg News previously reported — a significant rise from the roughly $50 billion valuation DeepSeek achieved in its first round.

The company has also begun preparations for an initial public offering and may file as soon as this year, potentially marking a landmark debut for China's technology sector and one of the most high-profile AI listings globally.

DeepSeek has attracted intense interest from prospective investors because it occupies a central position in China's push to compete globally on artificial intelligence. The company developed a model last year that stunned the industry by demonstrating the ability to build a cutting-edge yet efficient AI platform using fewer computing resources. The breakthrough showed that Chinese firms could rival the top players in Silicon Valley despite U.S. export restrictions on advanced hardware, and it intensified the broader debate over whether export controls alone can constrain China's AI progress.

The startup is now seeking additional capital to support an ambitious expansion plan, including increased computing capacity. AI laboratories worldwide are pursuing deals to secure the data center infrastructure required to train and operate their services, and access to sufficient compute has become a defining competitive factor across the industry.

Founded in 2023, DeepSeek is owned by hedge fund Zhejiang High-Flyer Asset Management. Its initial fundraising round set a record for first-time financing by a Chinese technology startup. Beyond Tencent and CATL, it also secured backing from the National Artificial Intelligence Industry Investment Fund, one of the key vehicles driving Beijing's overarching strategy in the sector.

DeepSeek's senior management has told potential investors that the startup intends to prioritize groundbreaking AI research over short-term commercialization, Bloomberg News has reported. In at least one meeting with investors, Liang pledged to continue developing open-source AI models while pursuing the broader objective of achieving artificial general intelligence, reinforcing the company's emphasis on advancing the frontiers of AI rather than focusing on monetization.

Liang's net worth more than doubled following the firm's most recent fundraising round, making the Chinese entrepreneur the world's wealthiest founder of an AI model company. He is now worth $36 billion, up from approximately $16.7 billion previously, according to the Bloomberg Billionaires Index. That places him well above Anthropic PBC co-founder Dario Amodei and OpenAI's Greg Brockman.