Key players: Anthropic (AI firm behind Claude, first deployed on Pentagon classified networks, with up to $200M DoD contract); U.S. Pentagon/Department of War (led by Secretary Pete Hegseth, spokesperson Sean Parnell, negotiator Defense Undersecretary Emil Michael); President Trump; rivals OpenAI (recently agreed to deploy models on classified networks) and mentions of Google/xAI.[1][2][3][5] Legal experts like Dean Ball and Amos Toh question the designation's legality, citing missing risk assessments and congressional notification.[4]
Context and timeline: Dispute arose from months of failed negotiations where Pentagon sought unrestricted use of Claude for "all lawful purposes," including potential mass surveillance (e.g., geolocation, browsing data) and autonomous weapons; Anthropic refused safeguards removal due to safety/reliability concerns.[1][2][5] Tensions escalated mid-February with threats of risk label; Friday deadline (Feb 27, 5:01 PM) unmet, prompting Hegseth's X declaration and Trump's blacklist, unprecedented for a U.S. firm (typically for adversaries like Huawei).[3][4][5]
Newsworthy now: Marks first U.S. "supply chain risk" on an American company, allegedly in retaliation for contract terms, risking broad fallout for Anthropic's $14B revenue business (used by 8/10 largest U.S. firms) and setting precedent amid AI-national security clashes; coincides with OpenAI's Pentagon deal, highlighting policy shifts under Trump/Hegseth.[1][3][4][5]