DeepSeek AI models, banned by the US government

DeepSeek AI Models Banned in the US Government The US government has officially prohibited its agencies from acquiring and using artificial intelligence (AI) models from the Chinese company DeepSeek, which has been accused of patent infringement by its American competitors and the administration of President Donald Trump, according to AFP. The decision was issued by the Department of Homeland Security (DHS), which has authority over civil government agencies but not over the military or US intelligence services. The US Department of Defense did not respond to a request for comment from AFP. In January 2025, the US Navy banned the internal use of DeepSeek. The Department of Homeland Security did not provide a justification for its decision in the statement. Some have expressed concern about the potential cybersecurity risks associated with Chinese AI models, but no incidents have been reported to date. Furthermore, most of the top artificial intelligence systems in China are open source, meaning they can be downloaded and modified, thus escaping the control of their developers, according to Agerpres. OpenAI and Anthropic have repeatedly accused DeepSeek of plagiarizing models using a process known as distillation. This involves sending a large number of requests to an artificial intelligence and then analyzing the results to imitate and develop their own interface. At the beginning of September, in a document published jointly by the Cybersecurity and Infrastructure Security Agency (CISA), the National Security Agency (NSA), and the FBI, the US government presented a detailed description of the methods used by major Chinese players in the field of artificial intelligence to bypass the most advanced American models. Donald Trump considers the development of this technology a race between the United States and China. So far, the Trump administration's policy has mainly focused on limiting regulatory constraints imposed on large American AI companies and encouraging investment in this technology, without attempting to restrict or ban Chinese models in the United States. Chinese artificial intelligence is becoming increasingly popular in the West as a low-cost alternative to the main American models, which are considered closed systems, meaning they cannot be modified.