The US Department of Homeland Security has ordered all artificial intelligence technology from Chinese firm DeepSeek removed from most government systems, banning its use within civilian agencies along with any technology built using its AI models.

The order was made public on Monday but took effect at the end of July. It applies to agencies including the Treasury, State and Justice departments, though Department of Defense agencies are excluded. The DHS notice did not spell out the specific security concerns around DeepSeek.

The ban follows criticism of Chinese AI from the Trump administration and from American companies. White House tech lead Michael Kratsios wrote in an internal memo earlier this year that "malicious exploitation" of American AI firms had been occurring, mainly by rivals in China. Anthropic and OpenAI have also claimed some Chinese AI firms, including DeepSeek, essentially copied their models.

DeepSeek's open source models have been seen as a disruptive force in the AI race, costing developers a fraction of closed alternatives. The company said training the debut version of its model took about $6m, against roughly $100m for OpenAI's GPT-4 - a disclosure that sent tech stocks tumbling. Its latest version was released last week, with users praising its performance on tasks such as software coding.

The DHS, DeepSeek and the Chinese embassy in Washington DC have been contacted for comment.

Source: BBC News