Recently, CNBC learned that the US Navy has instructed its forces to avoid using the artificial intelligence technology named DeepSeek developed by China (CCP).
The US Navy issued a warning to its forces via email on Friday, January 24, citing potential security and ethical issues related to the source and use of the DeepSeek artificial intelligence, stating that it is “imperative” not to use it in any form for work-related tasks or personal use.
The email stated, “We want to bring your attention to a significant update of a new AI model called DeepSeek … DeepSeek’s AI should not be used for any work-related tasks or personal use,” emphasizing its urgency.
Recipients were instructed not to download, install, or use the DeepSeek model in any form.
A Navy spokesperson confirmed the authenticity of the email, stating that it pertains to the AI policy established by the Navy’s Chief Information Officer.
The Navy’s warning came after DeepSeek released its new reasoning AI model, R1.
The email was sent to the OpNav (Naval Operations) distribution list on Friday morning, indicating it was a memorandum for all, approved by the Naval Air Warfare Center Division Cyber Workforce Manager.
Industry experts have highly praised DeepSeek’s performance and reasoning abilities, comparing the model favorably with OpenAI’s technology. The open-source DeepSeek model is available for use by any AI developer. The application has quickly risen to the top spot in the Apple App Store, displacing OpenAI’s ChatGPT.
CEO of Scale AI, Alexandr Wang, told CNBC last week that DeepSeek’s latest AI model has “disrupted the world,” with its R1 version even more powerful, on par with the best models America has to offer. He described the competition between the US and China as an “AI war.”
His company provides training data for companies like OpenAI, Google, and Meta.
In December, DeepSeek announced that despite the US restricting chip exports to China three times in three years, their large language model was developed in just two months at a cost of less than $6 million, only a fraction of what OpenAI, Anthropic, Google, and other companies spent.
DeepSeek shocked the capital market on Monday, with investors speculating that the infrastructure needed for future AI products may be cheaper than Wall Street anticipated, expressing concerns that low-cost Chinese AI models could threaten the market dominance of American leaders like OpenAI and Google.
Shares of American AI chip manufacturers NVIDIA and Broadcom fell by 17% on Monday, resulting in a combined $800 billion loss in market value. These stocks contributed to a 3.1% decline in the Nasdaq index.
Last week, President Trump took a significant step in the AI field by announcing the joint venture “Stargate” between OpenAI, Oracle, and SoftBank, investing billions of dollars in AI infrastructure in the US.
President Trump stated that DeepSeek’s sudden rise has sounded an alarm for American tech companies, but he expressed confidence that Beijing will not dominate the AI field.
He said, “We will dominate, we will lead everything.”
David Sacks, Trump’s AI and cryptocurrency advisor and venture capitalist, posted on X on Monday that DeepSeek R1 indicates the “AI race will be fierce.”
He added, “I am confident in America, but we cannot be complacent.”
On Tuesday, White House spokesperson Karoline Leavitt stated that US officials are reviewing the national security implications of the Chinese AI application DeepSeek. Meanwhile, the Italian government is requesting DeepSeek to explain its collection and use of personal data.
Leavitt, in her first briefing as White House spokesperson on Tuesday, mentioned that the National Security Council is reviewing the impact of DeepSeek’s application, calling it a warning for the American AI industry.
She added that she had personally discussed the matter with the National Security Council earlier in the day, emphasizing the White House’s efforts to “ensure US dominance in AI.”
The House Select Committee on China urged strengthened export controls on the technology behind DeepSeek’s AI infrastructure, stating, “The US cannot allow CCP models like DeepSeek to jeopardize our national security and exploit our technology for its AI ambitions.”