Anthropic Sues to Stop Pentagon AI Restrictions

Anthropic Challenges Pentagon's National Security Designation

Anthropic, an artificial intelligence laboratory, has taken legal action to prevent the Pentagon from placing it on a national security blacklist. This move marks a significant escalation in the company’s ongoing conflict with the U.S. military over restrictions on its technology. In a lawsuit filed in federal court in California, Anthropic argued that the designation was unlawful and violated its free speech and due process rights. The company is seeking a judge's decision to overturn the designation and halt any enforcement by federal agencies.

"These actions are unprecedented and unlawful. The Constitution does not allow the government to wield its enormous power to punish a company for its protected speech," Anthropic stated in the filing.

The Pentagon designated Anthropic as a supply-chain risk on Thursday, which limits the use of its technology in military operations. According to two sources, this technology has been used for military activities in Iran. Defense Secretary Pete Hegseth made the designation after the startup refused to remove guardrails against using its AI for autonomous weapons or domestic surveillance.

The Broader Implications of the Dispute

The dispute between Anthropic and the U.S. government highlights a larger issue about the balance of power between the administration and companies developing AI. The fight could set a precedent for how other AI companies negotiate restrictions on the military use of their technology.

Anthropic has long sought to engage with the U.S. national security apparatus, even before many other AI companies did. CEO Dario Amodei has expressed no opposition to AI-driven weapons but believes current AI technology lacks the accuracy required for such applications. The company has indicated that the lawsuit does not rule out the possibility of re-opening negotiations with the U.S. government and reaching a settlement. However, the Pentagon has stated it will not comment on litigation, and a recent statement from a Pentagon official suggested that active talks have ceased.

Financial and Reputational Risks

The designation poses a significant threat to Anthropic’s business with the government, potentially affecting its revenue and reputation. Analysts have warned that the outcome could impact the company's enterprise operations, as some enterprises might pause deployments of its tools while the legal battle unfolds.

"The government's actions immediately and irreparably harm Anthropic," said Thiyagu Ramasmy, Head of Public Sector at Anthropic. Krishna Rao, the finance chief, added that if the government's actions stand, the impact on the company would be "almost impossible to reverse."

In specific examples, Chief Commercial Officer Paul Smith noted that a partner with a multi-million-dollar annual contract had switched from Claude to a rival generative AI model, eliminating an anticipated revenue pipeline of more than $100 million. Additionally, negotiations with financial institutions worth roughly $180 million combined have been disrupted.

Supply-Chain Risk Designation

In addition to the initial lawsuit, Anthropic filed a second lawsuit on Monday, challenging the government's designation of it as a supply-chain risk under a broader law. This designation could lead to a blacklisting across the entire civilian government. The scope of this designation remains unclear, as the government must conduct an interagency review to determine the extent of the restrictions.

A group of 37 researchers and engineers from OpenAI and Google filed an amicus brief in support of Anthropic, arguing that the episode could discourage AI experts from openly debating AI's risks and benefits. The group, including Google Chief Scientist Jeff Dean, emphasized that silencing one lab could reduce the industry's potential to innovate solutions.

Political and Industry Reactions

The Pentagon's actions came after months of discussions with Anthropic regarding whether the company's policies could constrain military action. Shortly after meeting with Hegseth, the Pentagon announced on February 27 that it would declare Anthropic as a supply-chain risk. It officially informed Anthropic of the designation on March 3.

The Pentagon has maintained that U.S. law, not private companies, should determine how to defend the country and insisted on having full flexibility in using AI for "any lawful use." Anthropic, however, argues that even the best AI models are not reliable enough for fully autonomous weapons and that using them for such purposes would be dangerous. The company also drew a red line on domestic surveillance of Americans, calling that a violation of fundamental rights.

After Hegseth's announcement, Anthropic stated that the designation would be legally unsound and set a dangerous precedent for companies that negotiate with the government. The company reiterated its commitment to challenging the designation in court.