The technology war between the United States and China has now extended into the realm of artificial intelligence (AI). The Department of Homeland Security (DHS) has prohibited government agencies from using the DeepSeek AI model; although this policy has been in place since July, it was only publicly announced yesterday.
Government agencies have been instructed to immediately remove any DeepSeek-based systems from their operations. DeepSeek models were previously favored due to their open-source nature and lower costs compared to models offered by American companies. The Department of Defense is the only entity permitted to continue using DeepSeek—likely to monitor the capabilities of AI models developed by China.
Launched last year, DeepSeek became a phenomenon in the AI world, offering high performance at a significantly lower training cost. Taiwan, South Korea, and Australia were among the first nations to ban the use of DeepSeek on government-owned devices, citing potential security risks.
Last year, Senator Josh Hawley also proposed measures to prevent U.S. citizens from assisting in the advancement of the People's Republic of China’s AI capabilities. Those found guilty of utilizing AI developed by China could face penalties including up to 20 years in prison, fines of up to $1 million for individuals, or fines of up to $100 million for corporations.
