Microsoft Vice Chairman and President Brad Smith told a US Senate hearing that the company banned its employees from using the DeepSeek app due to data security concerns and concerns about the spread of Chinese propaganda. About this informs TechCrunch.
“At Microsoft, we do not allow our employees to use the DeepSeek application”, – noted Smith.
Also, DeepSeek is not included in Microsoft’s application catalog due to security concerns. Although some organizations and countries have already taken restrictive measures against this application, Microsoft has publicly announced its ban for the first time.
According to Smith, the restrictions are due to the fact that user data is stored on Chinese servers that fall under the jurisdiction of the PRC, which requires companies to cooperate with national intelligence agencies. In addition, DeepSeek is reportedly heavily censoring information on topics the Chinese government considers sensitive.
Despite these risks, Microsoft recently integrated the DeepSeek R1 model into its Azure cloud service. Thanks to the open source, anyone can download the model, save it locally and work with it without transferring the data to China. However, other dangers remain, including the possibility of spreading propaganda or creating malicious code.
Smith noted during the hearing that Microsoft was able to “penetrate” DeepSeek’s artificial intelligence model and eliminate side effects, although the company did not specify exactly what changes were made. As part of the launch of DeepSeek on Azure, Microsoft said that the model has undergone an in-depth security review.
We will remind that in April 2025, a special committee of the US House of Representatives recognized DeepSeek as a threat to national security and called on the US government to strengthen export controls and minimize strategic risks associated with advanced Chinese AI models.
