Microsoft staff barred from using DeepSeek app by company order, president confirms
Microsoft prohibits its employees from using DeepSeek over data security and content moderation concerns, according to Brad Smith, Microsoft's vice chairman and president, during a Senate hearing.
"Microsoft restricts employee access to DeepSeek's application service," Smith stated, noting its availability across desktop and mobile platforms.
Smith confirmed Microsoft hasn't approved DeepSeek for its app store due to these security considerations.
While numerous organizations and governments have limited DeepSeek usage, this marks Microsoft's first public acknowledgment of such restrictions.
The executive cited risks including Chinese data storage requirements and potential influence from government-aligned content filtering.
DeepSeek's privacy policy confirms data storage on Chinese servers under local jurisdiction, which includes mandatory cooperation with intelligence agencies and strict censorship protocols.
Despite these criticisms, Microsoft partnered with DeepSeek to offer its R1 model through Azure cloud services following the model's viral success.
Showcase Your AI Innovation at TechCrunch Sessions: AI
Reserve your exhibition space at TC Sessions: AI to present your technology to 1,200+ industry leaders—affordable opportunities available until May 9 or until spots fill.
Showcase Your AI Innovation at TechCrunch Sessions: AI
Reserve your exhibition space at TC Sessions: AI to present your technology to 1,200+ industry leaders—affordable opportunities available until May 9 or until spots fill.
The Azure partnership differs from distributing DeepSeek's chatbot directly. As an open-source model, organizations can deploy DeepSeek locally without Chinese data routing.
However, risks remain regarding potential algorithmic biases and security vulnerabilities in generated outputs.
Smith testified that Microsoft modified DeepSeek's model to eliminate "harmful outputs," though specifics weren't disclosed beyond referring to Senate testimony.
Microsoft's Azure launch documentation highlighted DeepSeek's extensive security testing prior to platform integration.
While DeepSeek competes with Microsoft's Copilot, the Windows store hosts other AI competitors like Perplexity—though notably excludes Google's Gemini and Chrome applications.
Related article
DeepSeek-V3 Unveiled: How Hardware-Aware AI Design Slashes Costs and Boosts Performance
DeepSeek-V3: A Cost-Efficient Leap in AI DevelopmentThe AI industry is at a crossroads. While large language models (LLMs) grow more powerful, their computational demands have skyrocketed, making cutting-edge AI development prohibitively expensive for most organizations. DeepSeek-V3 challenges this
DeepSeek-GRM: Revolutionizing Scalable, Cost-Efficient AI for Businesses
If you're running a business, you know how tough it can be to integrate Artificial Intelligence (AI) into your operations. The high costs and technical complexity often put advance
New Technique Enables DeepSeek and Other Models to Respond to Sensitive Queries
Removing bias and censorship from large language models (LLMs) like China's DeepSeek is a complex challenge that has caught the attention of U.S. policymakers and business leaders, who see it as a potential national security threat. A recent report from a U.S. Congress select committee labeled DeepS
Comments (0)
0/200
Microsoft prohibits its employees from using DeepSeek over data security and content moderation concerns, according to Brad Smith, Microsoft's vice chairman and president, during a Senate hearing.
"Microsoft restricts employee access to DeepSeek's application service," Smith stated, noting its availability across desktop and mobile platforms.
Smith confirmed Microsoft hasn't approved DeepSeek for its app store due to these security considerations.
While numerous organizations and governments have limited DeepSeek usage, this marks Microsoft's first public acknowledgment of such restrictions.
The executive cited risks including Chinese data storage requirements and potential influence from government-aligned content filtering.
DeepSeek's privacy policy confirms data storage on Chinese servers under local jurisdiction, which includes mandatory cooperation with intelligence agencies and strict censorship protocols.
Despite these criticisms, Microsoft partnered with DeepSeek to offer its R1 model through Azure cloud services following the model's viral success.
Showcase Your AI Innovation at TechCrunch Sessions: AI
Reserve your exhibition space at TC Sessions: AI to present your technology to 1,200+ industry leaders—affordable opportunities available until May 9 or until spots fill.
Showcase Your AI Innovation at TechCrunch Sessions: AI
Reserve your exhibition space at TC Sessions: AI to present your technology to 1,200+ industry leaders—affordable opportunities available until May 9 or until spots fill.
The Azure partnership differs from distributing DeepSeek's chatbot directly. As an open-source model, organizations can deploy DeepSeek locally without Chinese data routing.
However, risks remain regarding potential algorithmic biases and security vulnerabilities in generated outputs.
Smith testified that Microsoft modified DeepSeek's model to eliminate "harmful outputs," though specifics weren't disclosed beyond referring to Senate testimony.
Microsoft's Azure launch documentation highlighted DeepSeek's extensive security testing prior to platform integration.
While DeepSeek competes with Microsoft's Copilot, the Windows store hosts other AI competitors like Perplexity—though notably excludes Google's Gemini and Chrome applications.
DeepSeek-V3 Unveiled: How Hardware-Aware AI Design Slashes Costs and Boosts Performance
DeepSeek-V3: A Cost-Efficient Leap in AI DevelopmentThe AI industry is at a crossroads. While large language models (LLMs) grow more powerful, their computational demands have skyrocketed, making cutting-edge AI development prohibitively expensive for most organizations. DeepSeek-V3 challenges this
DeepSeek-GRM: Revolutionizing Scalable, Cost-Efficient AI for Businesses
If you're running a business, you know how tough it can be to integrate Artificial Intelligence (AI) into your operations. The high costs and technical complexity often put advance
New Technique Enables DeepSeek and Other Models to Respond to Sensitive Queries
Removing bias and censorship from large language models (LLMs) like China's DeepSeek is a complex challenge that has caught the attention of U.S. policymakers and business leaders, who see it as a potential national security threat. A recent report from a U.S. Congress select committee labeled DeepS
