Anthropic CEO Dario Amodei said Thursday that the artificial intelligence company “cannot in good conscience agree” to the Pentagon’s request for unrestricted use of its technology, escalating an unusually public standoff with Donald Trump’s administration that could cost the company government contracts as early as Friday.The company behind the artificial intelligence chatbot Cloud said it remained open to talks but warned that the Defense Department’s revised contract language “makes little progress in preventing Cloud from being used for mass surveillance of Americans or fully autonomous weapons,” according to the Associated Press.Chief Pentagon spokesman Sean Parnell dismissed those concerns, writing on social media that the military “has no interest in using AI to conduct mass surveillance of Americans (which is illegal), nor do we want to use AI to develop autonomous weapons that require no human involvement.”Anthropic’s internal policy prohibits such use. The company is currently the only major AI developer that has not agreed to provide its technology to the new internal U.S. military network – joining the likes of Google, OpenAI and Elon Musk’s xAI.“The department has the right to select the contractor that best fits their vision,” Amodei said in a statement. “But given the tremendous value Anthropic’s technology provides to our armed forces, we hope they reconsider.”Controversy further intensified after Defense Minister took office Peter Heggs Meeting with Amodei on Tuesday, he issued an ultimatum: Allow unrestricted use of Anthropic’s artificial intelligence technology by Friday or risk losing the Pentagon contract. Officials also warned that the company could be designated a supply chain risk or that the Cold War-era Defense Production Act could be invoked to give the military broader authority over its products.Amodei criticized the threats as inconsistent, saying “the latter two threats are inherently contradictory: one labels us a security risk; the other labels Crowder critical to national security.”Parnell reiterated that the Pentagon seeks to “use Anthropic’s models for all lawful purposes,” but did not specify what those uses include. He argued that wider access to the technology was necessary to avoid “jeopardizing critical military operations.”“We will not let any company dictate the terms of how we make operational decisions,” he said.Negotiations between the two sides have been ongoing for several months. Amodai said Anthropic “will work toward a smooth transition to another provider” if the Pentagon does not change its stance.The public nature of the dispute has drawn criticism from Capitol Hill.North Carolina Republican Sen. Thom Tillis said the Pentagon’s handling of the issue was unprofessional and said Anthropic was “trying to help us.”“Why on earth are we discussing this publicly?” Tillis told reporters. “That’s not the way to deal with strategic suppliers that you have contracts with.”He added, “When a company turns down a market opportunity because of fear of negative consequences, you should listen to them and then privately figure out what problem they are really trying to solve.”Virginia Sen. Mark Warner, the top Democrat on the Senate Intelligence Committee, said he was “deeply troubled” by reports that the Pentagon is “committed to bullying America’s leading nations.” company. ““Unfortunately, this is further evidence that the Department of Defense is trying to completely ignore AI governance,” Warner said. It “further emphasizes the need for Congress to enact strong, binding AI governance mechanisms for national security.”Pentagon officials insist that artificial intelligence systems will be used in accordance with the law, even as the department has sought to reshape its internal legal culture.Hegseth told Fox News last February that the military wanted lawyers to provide constitutional advice but not be an “obstacle.” That same month, he fired top legal officials from the Army and Air Force without explanation. The Navy’s top lawyer resigned shortly after the 2024 election.
