Trump Administration Bans DeepSeek Across Federal Agencies
Homeland Security mandates removal of Chinese AI tools citing data privacy and security risks.
1Federal Ban Implementation
The Department of Homeland Security issued a formal notice requiring federal civilian agencies to remove DeepSeek technology. This directive encompasses both direct product usage and third-party services reliant on DeepSeek AI models. Agencies like the Department of State must now purge these tools to comply with the new administration-wide security mandate.
2Data Privacy Concerns
Homeland Security Secretary Markwayne Mullin highlighted that DeepSeek’s account generation process harvests personally identifiable information. The administration fears this data could be transferred to the Chinese government. Furthermore, officials warned that the models themselves might contain embedded malware, posing a direct threat to the integrity of sensitive federal government digital infrastructure.
3Escalating Security Threats
The U.S. government has officially labeled DeepSeek a profound threat to national security. Prior to this broad ban, the Navy and the Department of Commerce had already restricted internal access. These measures follow the release of the R1 model, which disrupted the American AI industry by offering high capabilities at lower costs.
4The Distillation Controversy
American AI giants, including Anthropic, have accused DeepSeek of illicitly using model distillation to train their systems. This process involves using outputs from advanced teacher models to train student models. The U.S. government now views these specific distillation attempts as adversarial, effectively equating the practice to industrial-scale intellectual property theft and cheating.
5Regulatory and Sanction Threats
Treasury Secretary Scott Bessent has threatened potential sanctions against Chinese AI companies involved in these distillation campaigns. A joint letter from the FBI, NSA, and CISA specifically named DeepSeek as a primary actor. Officials are now considering aggressive policy actions to be implemented in the coming days or weeks to curb this.
6Shifting Competitive Landscape
While American AI models maintain a lead, the gap is narrowing as Chinese labs like DeepSeek and Moonshot accelerate development. The release of the R1 model served as a significant signal of this catch-up. These advancements have intensified debates in Washington regarding the long-term national security implications of losing the global AI race.
7Beijing’s Internal Oversight
Interestingly, Beijing is also expressing concerns regarding the national security implications of its own AI sector. Authorities have recently proposed measures that could impose export controls on the overseas deployment of certain Chinese AI models. This suggests that China is attempting to keep its most advanced AI labs on a tighter leash.
8Strategic Diplomatic Context
The U.S. and China are scheduled to engage in talks regarding AI security through the US-China AI Dialogue framework. This upcoming diplomatic engagement likely explains the timing of the current crackdown. The administration is balancing the need for international cooperation with the urgent requirement to fortify domestic defenses against potential foreign intelligence threats.
Original article: Gizmodo — Trump Administration Bans DeepSeek Across Federal Agencies