Anthropic Officially Responds: Will Sue the U.S. Department of Defense Over "Supply Chain Risk" Determination

The Tong Finance APP has learned that the U.S. Department of Defense has officially notified Anthropic PBC that the company and its products pose a risk to the U.S. supply chain. According to a senior defense official, this move further intensifies the dispute between the two parties over AI security.

The official revealed on Thursday: “The Department of War (DOW) has formally notified Anthropic’s leadership that the company and its products are considered supply chain risks, and this designation is effective immediately.” Notably, Defense Secretary Pete Hegseth has recently preferred to use the traditional term “Department of War” to refer to the Department of Defense, and this is the first time the official statement has been issued under that name.

Facing the threat of a blacklisting issued by the Department of Defense over AI security concerns, Anthropic PBC expects this deadlock to escalate into a legal confrontation.

“We firmly believe this action lacks legal basis, so we have no choice but to defend ourselves through judicial channels,” Anthropic CEO Dario Amodei stated clearly on Thursday in an official blog.

Although the defense official emphasized that the decision “takes effect immediately,” sources familiar with the matter revealed that the U.S. military continues to use Anthropic’s Claude AI tools in operations against Iran. Last Friday, when Secretary Hegseth issued a warning to the company, he outlined a six-month transition period, requiring the transfer of AI operations to other vendors.

Despite the official’s statement that the decision “takes effect immediately,” sources said that the U.S. military is still actively using Anthropic’s Claude AI tools in its operations against Iran. Hegseth’s warning last Friday included a six-month transition window for transferring AI activities to other providers.

Both Anthropic and the U.S. Department of Defense did not immediately comment on this series of developments. Notably, the defense official did not specify when or how the Department of Defense communicated this risk designation to Anthropic.

Previously, Anthropic had explicitly stated that it would take legal action against any supply chain risk designation made by the Department of Defense.

The Department of Defense’s risk assessment could disrupt the collaboration between Anthropic and the military—an area where the company has long been deeply dependent on its software. Until recently, Anthropic remained the only approved AI system provider on the U.S. Department of Defense’s classified cloud platform, with its Claude Gov tool becoming a core operational platform for defense personnel due to its exceptional usability.

“This represents a highly strategic technological capability,” said Lauren Kahn, senior research analyst at the Center for Security and Emerging Technology at Georgetown University. “If this capability is forcibly stripped away, it could trigger a chain reaction of damage across the military, industry, and the entire tech ecosystem.”

Dario Amodei, CEO of Anthropic, had engaged in weeks of negotiations with Deputy Secretary of Defense for Research and Engineering Emil Michael to finalize a specialized contract governing how the U.S. Department of Defense can access Anthropic’s technology. However, the negotiations broke down last week—this startup insisted on clear guarantees that its AI technology would not be used for mass surveillance of U.S. citizens or autonomous weapons deployment.

Subsequently, on Friday, Secretary Hegseth posted on X (formerly Twitter) that Anthropic had become a “supply chain risk.” It is noteworthy that such risk designations typically target countries or entities considered adversaries by the U.S., making this application to a tech company quite rare.

It remains unclear under which law the U.S. Department of Defense is classifying Anthropic as a supply chain threat. In a statement responding to Hegseth’s social media post last week, Anthropic indicated that it expects the risk designation to ultimately be executed under Section 3252 of the U.S. Armed Forces Management Act.

“From the very beginning, the core principle has been clear— the military must have the autonomy to use technology for all lawful purposes,” the defense official emphasized Thursday. “We will never allow any supplier to interfere with command chains by restricting the lawful use of critical capabilities, thereby endangering our personnel.”

This designation comes at a critical point when U.S. forces are heavily reliant on Claude tools in their operations against Iran—using a series of AI tools to efficiently process vast amounts of operational data. Sources familiar with the matter revealed that the Maven intelligence system, developed by Palantir Technologies Inc., is widely used in the Middle East, and Anthropic’s Claude AI is one of the core large language models integrated into this system. These sources stressed that Claude has performed excellently in real-world applications, becoming a vital support for U.S. actions against Iran and significantly accelerating AI upgrades within the Maven system.

Currently, Anthropic’s valuation has reached approximately $380 billion. Based on current performance, the company is projected to achieve nearly $20 billion in annual revenue—doubling its revenue run rate compared to the end of last year. However, ongoing disputes with the U.S. Department of Defense cast a shadow over the company’s future prospects.

The long-term impact of the U.S. Department of Defense’s risk designation on Anthropic’s enterprise sales—its core revenue pillar—remains to be seen. Notably, amid the dispute, the company is quietly expanding into the consumer market—its main app recently topped the Apple App Store download charts, reflecting broad market recognition and user support for Anthropic’s technology products.