Anthropic Faces Pentagon Scrutiny
AI firm Anthropic is in hot water with the Trump administration after a critical memo leaked, prompting a public CEO apology. The Pentagon is now considering designating the company a "supply chain risk," which could restrict its ability to win defense contracts. Partner Microsoft has already pledged to wall off Anthropic's tech from any Pentagon use.
The critical memo from CEO Dario Amodei alleged the Trump administration's hostility stemmed from Anthropic's refusal to donate to the president's campaign or offer "dictator-style praise," unlike some competitors. This followed a history of friction, including Anthropic's lobbying against a Trump-backed bill that aimed to pause state-level AI regulations for a decade. The "supply chain risk" designation is a first for a U.S. company; this label is typically reserved for foreign-based firms, such as China's Huawei, suspected of being influenced by adversarial governments. By law, the designation is meant to prevent adversaries from sabotaging or subverting critical systems. The conflict's core is Anthropic's refusal to remove ethical "red lines" that prevent its Claude AI models from being used for mass domestic surveillance or fully autonomous weapons systems. The Pentagon, under Defense Secretary Pete Hegseth, has demanded "full, unrestricted access" for all lawful military purposes. Anthropic intends to challenge the Pentagon's designation in court, arguing the action is not legally sound. CEO Dario Amodei stated the company sees no other option but to legally contest the designation, which he claims is being used to settle a policy dispute rather than address a genuine security threat. Despite the blacklisting, major cloud providers are standing by the company for non-defense work. Both Microsoft and Google have confirmed that Anthropic's AI models will remain available to commercial customers on their cloud platforms, including Microsoft's Azure and Google's Vertex AI. The designation could jeopardize Anthropic's recent $60 billion financing round and has already had significant consequences. The previous defense contract, signed in July 2025, was valued at $200 million. Major defense contractors like Boeing and Lockheed Martin must now cease using Anthropic's technology for any Pentagon-related projects. Hours after negotiations with Anthropic collapsed, the Pentagon announced a new deal with competitor OpenAI to use its technology on classified networks. The timing of the announcement drew criticism, with OpenAI's own CEO Sam Altman admitting his company's conduct appeared "opportunistic and sloppy."
