Pushback Mounts Over Anthropic's DoD 'Risk' Label
Tech workers and members of Congress are pushing back against the Trump administration's decision to label Anthropic a "supply chain risk" following failed contract negotiations. Rep. Sam Liccardo is reportedly preparing to introduce legislation to protect AI vendors, warning the move chills critical tech and defense partnerships.
The "supply chain risk" designation stems from failed negotiations where Anthropic sought to prohibit the use of its AI model, Claude, for mass domestic surveillance and fully autonomous weapons. The Pentagon, under Defense Secretary Pete Hegseth, insisted on contract language allowing for "any lawful use," leading to the impasse. This designation, typically used for foreign adversaries like Huawei, is unprecedented against a U.S.-headquartered company. It not only bars Anthropic from new Pentagon contracts but also prohibits any company doing business with the military from conducting commercial activity with Anthropic, a move with potentially vast ripple effects. Amazon and Google, both major Pentagon contractors, are also key cloud infrastructure providers for Anthropic. Anthropic has vowed to challenge the designation in court, calling it "legally unsound" and arguing the authority under 10 U.S.C. § 3252 doesn't extend to contractors' private commercial activities. The statute was designed to prevent adversaries from inserting vulnerabilities into sensitive military systems, not to resolve contract disputes with domestic firms. The administration's actions have been described as contradictory, with officials threatening to compel Anthropic's cooperation via the Defense Production Act one moment—implying the technology is vital—while simultaneously declaring it a national security risk that must be purged. President Trump also directed all federal agencies to cease using Anthropic's technology, with a six-month transition period for the DoD. The DoD's Chief Digital and Artificial Intelligence Office (CDAO) is central to the Pentagon's push to accelerate AI adoption and streamline procurement through initiatives like the Tradewinds marketplace. This office, formed in 2022, consolidated several AI and data-focused units to speed the deployment of capabilities for entities like Project Maven, the military's AI pathfinder for analyzing surveillance data. As the dispute with Anthropic unfolded, rival OpenAI announced it had reached a deal with the DoD. OpenAI stated its agreement included the same prohibitions on mass surveillance and autonomous weapons that Anthropic had sought, while also agreeing to the "any lawful use" standard.
