‘Extraordinary circumstances’: Pentagon could carves out exceptions to Anthropic AI ban amid legal battle
The US Pentagon has told its senior leaders that Anthropic may continue to use its AI tools beyond the six-month phase-out period, according to a report by Reuters. The report comes as hostilities between Anthropic and the US government are at their peak, with the Claude maker suing the Trump administration over what it describes as a supply chain risk.
Waivers to Anthropic may be granted in “rare and extraordinary circumstances” and “will only be considered for mission-critical activities that directly support national security operations where no viable alternative exists,” according to an internal March 6 memo signed by Pentagon Chief Information Officer Kirsten Davies and seen by Reuters.
Any Pentagon agency seeking to bypass the anthropic ban must submit a comprehensive risk mitigation plan for official approval.
The memo reportedly instructed officials to remove Anthropic products from systems that support critical missions such as nuclear weapons and ballistic missile defense.
It is also said to have been reaffirmed that the Anthropic ban also covers defense contractors. The memo gives Pentagon contracting officers 30 days to notify contractors and then document full compliance for up to a 180-day period.
Anthropic’s fight against Pentagon:
Meanwhile, Anthropic requested a stay of execution from the US appeals court after the Pentagon described it as a ‘supply chain risk’. The company claimed in court that this appointment could cost it billions of dollars.
“Under Anthropic’s best estimate for 2026, adverse government actions risk hundreds of millions, even billions, of dollars in lost revenue,” Anthropic’s attorney said in court.
“The federal government has retaliated against a leading AI developer for violating the U.S. Constitution and law by adhering to a protected viewpoint on an issue of great public importance, such as AI safety and the limitations of its AI model,” the company said. he said.
Meanwhile, the artificial intelligence initiative also received support from more than 30 OpenAI and Google employees who filed the lawsuit to support the initiative.
“The government’s designation of Anthropic as a supply chain risk was an inappropriate and arbitrary exercise of power with serious consequences for our industry,” the employees said in a brief.
Technology giant Microsoft also submitted its own motion to support Anthropic. The Pentagon’s Anthropic designation “will have negative consequences for the entire technology sector and American business,” the company said.
“This is not the time to put at risk the AI ecosystem that the Administration helps support,” he added.
