Unfolding Dispute: Anthropic Refutes Pentagon’s National Security Allegations
Anthropic, a prominent player in the AI industry, has taken legal action to refute the Pentagon’s assertion that the company poses an “unacceptable risk to national security.” Late on a recent Friday afternoon, the AI firm filed two affidavits in a California federal court, arguing that the government’s case was built on technical misunderstandings and allegations that were not made during the months of negotiations leading up to the dispute.
The Origin of the Dispute
The controversy began in late February when President Trump and Defense Secretary Pete Hegseth announced they were severing ties with Anthropic. The move came after the AI company refused to allow unlimited military use of its technology.
The affidavits were submitted together with Anthropic’s response letter in its lawsuit against the Department of Defense. The hearing is scheduled for Tuesday, March 24, before Judge Rita Lin in San Francisco. The individuals who filed the affidavits are Sarah Heck, Head of Policy at Anthropic, and Thiyagu Ramasamy, Head of Public Sector at the company.
The Affidavits: Anthropic’s Defense
Sarah Heck’s Affidavit
Heck, a former National Security Council official, was personally present at the February 24 meeting where CEO Dario Amodei met with Defense Minister Hegseth and Pentagon Undersecretary Emil Michael. In her affidavit, Heck highlights a key inaccuracy in the government documents: that Anthropic sought some form of approval role in military operations. She asserts that this claim is purely fictitious, stating, “At no time during Anthropic’s negotiations with the department did I or any other Anthropic employee indicate that the company wanted such a role.”
Heck also disputes the Pentagon’s concern that Anthropic might disable or modify its technology during operation, stating that this issue was never raised during negotiations. She claims it first appeared in government court filings, denying Anthropic an opportunity to respond.
Thiyagu Ramasamy’s Affidavit
Ramasamy, who brought his extensive experience in managing AI deployments for government clients to Anthropic, also filed a statement. His technical expertise is critical in refuting the government’s claim that Anthropic could possibly intervene in military operations by disabling the technology or altering its behavior. Ramasamy states this is technically impossible. Once the AI system is deployed to a government-backed “air gap” system run by a third-party contractor, Anthropic loses access to it entirely. No remote kill switch, backdoor, or mechanism to push unauthorized updates exists. Any kind of “operational veto” is pure fiction, he suggests.
In addition, Ramasamy also rebuts the government’s assertion that Anthropic’s employment of foreign nationals makes it a security risk. He points out that Anthropic employees have passed a U.S. government security clearance, the same background check required to access classified information.
Anthropic’s Legal Stand
Anthropic’s lawsuit argues that the supply chain risk designation – the first ever assigned to an American company – constitutes government retaliation for the company’s public views on AI safety, which violates the First Amendment. The government, however, has dismissed this claim in a 40-page filing, stating that Anthropic’s refusal to allow all lawful military uses of its technology is a business decision, not protected speech.
This ongoing dispute has stirred much conversation within the tech industry and beyond, as it explores the intersection of AI technology, governmental use, and national security. The court hearing scheduled for March 24 will undoubtedly shed more light on this complex issue.
For more details on this unfolding story, read the full report Here.
