China vows countermeasures after US bans DeepSeek AI from federal systems

Beijing protests the US prohibition on DeepSeek AI software, demanding the ban be lifted while maintaining its own blockade of Western AI services.

Threat Times Systems DeskOctober 21, 20262 min read

Share LinkedIn Reddit

Illustration: A sleek black AI server unit sits behind a translucent barrier etched with red warning stripes. — AI-generated illustration: Threat Times

Key points

DHS ordered civilian federal agencies to remove DeepSeek products and stop future purchases.
China's Commerce Ministry vows resolute countermeasures, calling the US action discriminatory.
The ban extends to any software built on DeepSeek’s open-source models.

The US Department of Homeland Security has mandated that civilian federal agencies remove all software from Chinese AI vendor DeepSeek from their IT environments. This directive, issued in October 2026, also prohibits the purchase of any new products from the company or any other vendors utilizing DeepSeek’s open-source models, according to reporting by The Decoder.

Washington cites national security concerns as the primary driver for this exclusion, specifically the risk that sensitive government data could flow to Beijing through the use of DeepSeek’s technology. The Department of Homeland Security’s order effectively bars the vendor from federal infrastructure, reflecting broader US efforts to limit the role of Chinese technology in critical systems.

Political friction and reciprocal blocks

Beijing has responded with sharp criticism, with the Foreign Ministry accusing the United States of stretching the definition of national security beyond reasonable limits. The Chinese Commerce Ministry has formally demanded that the US lift the ban and has vowed to implement resolute countermeasures in response to what it describes as discriminatory treatment.

As Threat Times reported on October 21, 2026, Beijing rejected US security claims and warned of retaliation following this directive. While China protests the exclusion of its domestic vendor, it simultaneously maintains a comprehensive block on major Western AI services, including OpenAI’s ChatGPT, Anthropic’s Claude, and Google’s Gemini, across its entire market.

Regulatory landscape and intelligence laws

China’s regulatory environment for AI is strictly controlled, requiring services to undergo review by the Cyberspace Administration and comply with political guidelines since 2023. This rule applies to the entire market, not just government entities, and state bodies are tasked with gradually replacing foreign hardware and software with domestic alternatives.

The US concerns are grounded in China’s 2017 National Intelligence Law, which legally requires companies and citizens to support intelligence services. Additionally, Chinese AI models are noted for steering clear of certain topics and repeating state propaganda with less critical scrutiny than their Western counterparts, according to The Decoder.

What should you do?

Organizations relying on AI tools should review their vendor contracts and data handling policies to ensure compliance with evolving federal directives. Security teams should monitor for updates regarding the scope of the DeepSeek ban and any new Chinese export controls on AI technologies that may impact global supply chains.

What to do and how to stay safe: DeepSeek

Audit current AI integrations to identify any use of DeepSeek or derivatives built on its open-source models.
Review data flow policies to ensure no sensitive information is being processed by unapproved foreign AI services.
Monitor official communications from the Department of Homeland Security for expanded lists of restricted vendors.
Evaluate contingency plans for replacing prohibited AI tools with approved domestic or allied alternatives.

Step-by-step guide: Prompt Injection Warning Signs: How to Spot AI Manipulation

General security guidance from the Threat Times newsroom. It is not confirmed advice from the organisations named in this story.