The Trump administration has ordered federal civilian agencies to remove artificial intelligence products developed by China’s DeepSeek from government information technology systems, citing serious national security and cybersecurity concerns. The Department of Homeland Security directive also prohibits future federal procurement of DeepSeek technology, including products developed by affiliated companies and third-party services incorporating its AI models. Federal officials warn that the technology could expose sensitive government information to Chinese authorities, facilitate intelligence gathering, or introduce cybersecurity vulnerabilities. Issued under the Federal Acquisition Supply Chain Security Act of 2018, the order represents a significant escalation in Washington’s efforts to protect federal information systems from potentially hostile foreign technology.
Key Takeaways
• Federal AI Ban: The Department of Homeland Security has ordered civilian federal agencies to eliminate DeepSeek technology from government systems and prohibit future purchases, including covered third-party applications.
• Chinese Intelligence Concerns: Federal officials identify potential transfers of sensitive information to China, intelligence exploitation, and embedded cybersecurity vulnerabilities as significant national security risks.
• Broader Technology Crackdown: The directive expands existing restrictions on Chinese artificial intelligence and underscores growing concerns about Beijing’s access to American government information and critical technology infrastructure.
In-Depth
The Trump administration has taken another significant step toward protecting American government technology from Chinese influence, ordering federal civilian agencies to remove artificial intelligence products developed by DeepSeek from their information systems.
The Department of Homeland Security issued the directive under the Federal Acquisition Supply Chain Security Act of 2018, following a review by the Federal Acquisition Security Council. The order prohibits agencies from purchasing DeepSeek products and requires removal of existing installations, including third-party applications incorporating the company’s AI technology.
Federal officials identified several potential threats, including unauthorized transfers of sensitive information to China, the exploitation of American government data for intelligence purposes, and cybersecurity vulnerabilities embedded within AI systems. These concerns reflect the broader national security challenges associated with technology developed under China’s regulatory and political environment.
DeepSeek has attracted international attention by producing comparatively inexpensive artificial intelligence models capable of competing with established American platforms. Its rapid advancement has intensified competition between Washington and Beijing, particularly as artificial intelligence becomes increasingly important to government operations, economic development, and national defense.
The restrictions follow earlier actions against DeepSeek by federal agencies and several state governments. However, the latest directive establishes broader protections across civilian federal information systems and procurement practices.
The decision also raises important questions about federal technology oversight. Agencies must identify not only applications purchased directly from Chinese developers but also software incorporating potentially vulnerable foreign AI components.
For an administration prioritizing American technological independence, the directive establishes a clear principle: federal efficiency and inexpensive software cannot justify exposing sensitive government information to potential foreign adversaries.
Sources
• https://public-inspection.federalregister.gov/2026-21447.pdf
• https://decrypt.co/381347/us-orders-civilian-agencies-rip-deepseek-out-government-systems
• https://thenextweb.com/news/us-bars-deepseek-federal-agencies