DeepSeek tools barred for US agencies over China data fears

The US Department of Homeland Security ordered civilian government agencies on Monday to stop using artificial intelligence tools from Chinese startup DeepSeek, citing concerns over data being transferred to China.

The restrictions cover products developed by DeepSeek and third-party tools created using DeepSeek. China's foreign ministry said it opposed the move and accused the United States of overstretching the concept of national security.

DeepSeek was founded by Liang Wenfeng in Hangzhou in 2023. It gained global attention in January 2025 with the release of its R1 deep-reasoning large language model, which it said was developed at a fraction of the cost of leading US systems.

DeepSeek's tools, like other Chinese chatbots, avoid topics usually censored in China, including the 1989 Tiananmen crackdown. Data privacy and censorship concerns have led several countries, including the United States, Australia and South Korea, to ban or restrict DeepSeek AI on government-issued devices.

DeepSeek's systems are open source, meaning developers can view and customise parts of the software. Similarweb says DeepSeek has about four percent of the global chatbot market, while ChatGPT has 68 percent.

The Information reported in December, citing six people with knowledge of the matter, that DeepSeek developed its V4 model using chips dismantled in third countries and smuggled to China despite US export controls. DeepSeek did not respond to a request for comment at the time. Nvidia told The Information it had not seen evidence of such smuggling and called the claim farfetched.