La Era
Technology

Anthropic Court Filing Challenges Pentagon National Security Designation Amid Contract Dispute

Anthropic challenges Pentagon's national security designation in federal court, citing sworn declarations from policy heads. The company disputes claims of operational risks and highlights discrepancies between private negotiations and public government statements. A hearing is scheduled for March 24 before Judge Rita Lin in San Francisco.

La Era

3 min read

Anthropic Court Filing Challenges Pentagon National Security Designation Amid Contract Dispute
Anthropic Court Filing Challenges Pentagon National Security Designation Amid Contract Dispute
Publicidad
Publicidad

Anthropic submitted two sworn declarations to a California federal court late Friday afternoon, directly challenging the Pentagon’s assertion that the AI company poses an unacceptable risk to national security. The company argues the government’s case relies on technical misunderstandings and claims that were never actually raised during the preceding months of negotiations. This legal maneuver comes ahead of a scheduled hearing this coming Tuesday, March 24, before Judge Rita Lin in San Francisco.

The dispute traces back to late February, when President Trump and Defense Secretary Pete Hegseth publicly declared they were cutting ties with Anthropic. This decision followed the company’s refusal to allow unrestricted military use of its AI technology in classified environments. The two individuals who submitted the declarations are Sarah Heck, Anthropic’s Head of Policy, and Thiyagu Ramasamy, the company’s Head of Public Sector.

Heck is a former National Security Council official who worked at the White House under the Obama administration before moving to private sector roles. She was personally present at the February 24 meeting where CEO Dario Amodei sat down with Defense Secretary Hegseth and the Pentagon’s Under Secretary Emil Michael. In her declaration, she calls out what she describes as a central falsehood in the government’s filings regarding operational approval roles.

According to Heck, the claim that Anthropic demanded approval over military operations is simply untrue during the negotiation period. She states that at no time did she or any other employee request that kind of role within the Department of Defense framework. She also claims the Pentagon’s concern about disabling technology mid-operation appeared for the first time in court filings, leaving the company no opportunity to respond.

A detail in Heck’s declaration draws significant attention regarding the timeline of the dispute between the two organizations. On March 4, the day after the Pentagon finalized its supply-chain risk designation, Under Secretary Michael emailed Amodei to say the sides were very close on key issues. This email appears alongside public statements where Michael claimed no active negotiations existed only days later.

Ramasamy brings different expertise to the case, having spent six years at Amazon Web Services managing AI deployments for government customers. His declaration takes on the government’s claim that Anthropic could theoretically interfere with military operations by disabling the technology remotely. He explains that once Claude is deployed inside a government-secured air-gapped system, Anthropic has no access to it.

Ramasamy states there is no remote kill switch or backdoor mechanism to push unauthorized updates into the secure environment. Any change to the model would require the Pentagon’s explicit approval and action to install physically or digitally. He further notes that Anthropic employees have undergone U.S. government security clearance vetting for access to classified information.

The lawsuit argues the supply-chain risk designation amounts to government retaliation for the company’s publicly stated views on AI safety. Anthropic claims this action violates the First Amendment by punishing protected speech rather than addressing actual security risks. The government rejected this framing entirely in a 40-page filing earlier this week.

The Pentagon asserts Anthropic’s refusal to allow all lawful military uses was a business decision rather than protected speech. They argue the designation was a straightforward national security call and not punishment for the company’s views on safety. The outcome of this hearing could define how private AI firms interact with defense contracting requirements globally.

Analysts suggest this case will set a precedent for future interactions between defense agencies and emerging AI vendors in the sector. The ruling may clarify the boundaries between national security designations and commercial speech rights for technology companies. Both sides will likely watch the March 24 hearing closely for procedural developments regarding the contract and potential appeals.

Publicidad
Publicidad

Comments

Comments are stored locally in your browser.

Publicidad
Publicidad