DeepSeek Considers New Funding Round After Raising $7 billion

DeepSeek Considers New Funding Round After Raising $7 billion

DeepSeek is reportedly considering a second financing round weeks after concluding its first.

The Chinese artificial intelligence company wrapped its first-ever round of funding near the end of May, raising $7 billion at a $52 billion valuation, the Financial Times (FT) reported Tuesday (July 14), citing three sources with knowledge of the matter.

DeepSeek this week began preliminary discussions with new investors about another funding round that would value the company at roughly $71 billion before the deal, two of the sources said. That figure would be a 37% increase in valuation, the report added, though details of the round have not yet been ironed out.

The sources added that DeepSeek’s unusually fast funding schedule is related to expectations of increased capital needed to construct its own data center and acquire more AI chips. The company’s efforts to develop AI agents is driving significantly greater demand for computing power, the report added.

According to the FT’s sources, DeepSeek founder Liang Wenfeng was the largest investor in the last round, putting around $3 billion of his own money into the company.

The report added that proceeds from the initial funding are being used to strengthen DeepSeek’s infrastructure and hire more AI researchers, amid greater competition in China and elsewhere to build the best models.

As the FT notes, DeepSeek emerged as a leader in AI research in China last year when it debuted its open-source R1 reasoning model, which performed at the same level as leading western systems but had been trained on more efficient methods.

In other AI news, Google DeepMind CEO Demis Hassabis on Tuesday proposed the creation of a U.S.-led standards body to independently test advanced artificial intelligence models for national security threats before release.

Artificial general intelligence that can match the human brain across a broad range of tasks is “probably only a few short years away,” leaving society a “precious window” to establish oversight, Hassabis said, according to a Financial Times report.

“The framework could give AI companies a common testing process while avoiding a separate licensing regime for every model. Independent testers could compare models against common thresholds and update tests as new capabilities emerge,” PYMNTS wrote.

“The White House has already moved toward pre-release oversight. A June executive order sought voluntary access to covered frontier models for up to 30 days before release to test advanced cyber capabilities. Google DeepMind, Microsoft and xAI previously agreed to provide models for federal national security testing.”