Microsoft prohibits Deeseek to its employees but offers Chinese AI to its customers

Deepseek

Deepseek is an emerging artificial intelligence platform that is positioned as a weight competitor in the face of other advanced technologies such as Chatgpt, Claude AI and Google Gemini.

• Downloads:
  1047
• Release date:
  04/30/2025
• Author :
  Deepseek
• License:
  Free license
• Categories:
  

  IA

• Operating system:
  

  Android, online service, iOS iPhone / iPad
  

“At Microsoft, we do not allow our employees to use the Deepseek application”said Brad Smith, both on PC and on mobile. The Redmond firm does not offer Chinese AI in the Microsoft Store for this reason, but its Azure customers have access to it. Remember that countries have already set up restrictions : This is the first time that Microsoft has publicly communicated on this ban.

Microsoft prohibits Deepseek to its employees

While Deepseek continues to improveBrad Smith talks about the risk that Microsoft employees data be stored in China and that the responses be influenced by “Chinese propaganda”.

Deepseek confirms in its privacy policy that user data is hosted on Chinese servers, and therefore subject to the laws of the country. AI also practices strong censorship on subjects considered to be sensitive in China.

Deepseek is offered to Azure customers

Despite these criticisms, and this is the most paradoxical, Microsoft proposed the R1 model of Deepseek on its Cloud Azure platform, Shortly after its launch which caused a stir. But the situation is different from the offer offered by the application. Deepseek is open source, so everyone can download the model, host it on their servers and offer it to its customers without the data being transmitted in China.

But the open source design does not prevent propaganda diffusion or the generation of malicious code. During his audition, Brad Smith explained that Microsoft changed the Deepseek AI to eliminate “harmful side effects”without developing on this subject. When launching on Azure, the Redmond firm said that the chatbot had undergone “rigorous safety assessments” before its deployment.