Democrats Launch Legislative Push After Pentagon Bans Anthropic AI Over Safety Dispute

Senate's Wyden Pledges Battle Over Pentagon Ban on Anthropic

Congressional Democrats are vowing to contest the Trump administration's unprecedented actions against the artificial intelligence giant Anthropic PBC, which include a prohibition on all use in the federal government and by Pentagon contractors. Ron Wyden of Oregon, the top Democrat on the Senate Finance Committee, vowed to "pull out all the stops" to fight back after Anthropic was barred by President Donald Trump from all government business, and said he plans to seek bipartisan backing for legislation to address the issue. Anthropic has said AI models were not yet reliable enough for autonomous weapons systems and also objected to using its technology to power mass surveillance of Americans, even if doing so might technically be legal. "I'm going to use every opportunity to try to undo the damage that he did to AI over the last couple of days," Wyden said of Trump. Earlier: Anthropic Made Pitch in Drone Swarm Contest During Pentagon Feud Representative Sam Liccardo, a California Democrat whose district includes a significant portion of Silicon Valley, said the debate over ground rules for deployment of AI technology shouldn't be happening behind closed doors between one company and one federal entity. "There's a reason why there ought to be a public discussion because it's going to impact all of us certainly when we're talking about like public surveillance," Liccardo said in an interview with Bloomberg Television on Monday. Although in the minority, Senate Democrats' votes will be needed to extend expiring surveillance and other authorities as well as to pass future defense spending and authorization bills. Wyden predicted some conservatives would also join the effort, given what he expected would be bipartisan concerns about widespread surveillance and allowing machines to make decisions on killing. "This is so breathtakingly wrong that I think we'll have support across the political spectrum," he said. Earlier: US Bars Anthropic Products From Agencies, Contractors Anthropic has already vowed to fight in court an order declaring it a supply chain risk barring all commercial work with companies contracting with the Pentagon, issued in a post by Defense Secretary Pete Hegseth on X Friday night in which he insisted that Trump, and not tech executives, will control the operational decisions of the military. "Effective immediately, no contractor, supplier, or partner that does business with the United States military may conduct any commercial activity with Anthropic," Hegseth said. Liccardo said he would be introducing a proposed addition to the Cold War-era law known as the Defense Production Act that would bar federal agencies from retaliating against companies like Anthropic for seeking reasonable limits on government use of their technology. "Technology is an obviously a wonderful thing, we know it has enormous risks we ought to take the cues from industry when industry tells us 'these risks are too great, let's draw lines,'" Liccardo added.

Dems tee up legislative response to Pentagon AI fight

Why it matters: The administration's unprecedented blacklisting of a leading American AI company could spur Congress to act on AI guardrails. Driving the news: Rep. Sam Liccardo (D-Calif.) plans to introduce an amendment to the Defense Production Act in response to the Anthropic-Pentagon dispute, his office tells Axios exclusively. * Liccardo's measure would amend the DPA to prohibit federal agencies from retaliating against AI and emerging tech vendors, per his office. * The DPA Modernization Act will be marked up on Wednesday by the House Financial Services Committee. * Liccardo represents Silicon Valley on Capitol Hill. Zoom in: While the Pentagon ultimately did not invoke the Defense Production Act in the Anthropic standoff, officials raised it as a possibility during negotiations to try to compel the company to remove its safety restrictions. * The Defense Production Act gives the president the authority to compel private companies to accept and prioritize particular contracts as required for national defense. * The DPA is set to expire on Sept. 30. What we're watching: Senate Democrats are in the early stages of weighing legislation related to the Anthropic-Pentagon battle, according to a source familiar with discussions. * They are considering legislation to broadly address concerns that the Pentagon is pursuing AI systems that could enable domestic surveillance or autonomous weapons without human oversight, per the source. * These discussions are in preliminary stages, but are expected to accelerate with lawmakers returning to the Hill on Monday. Flashback: Senate Armed Services Committee Chair Roger Wicker (R-Miss.) and Ranking Member Jack Reed (D-R.I.) along with Defense Appropriations Chair Mitch McConnell (R-Ky.) and Ranking Member Chris Coons (D-Del.) sent Anthropic and the Pentagon a private letter last week urging them to resolve the issue, the source said. * The bipartisan letter called for the Pentagon to extend its deadline, which it did not. The big picture: There are no comprehensive laws governing AI systems or consumer privacy in the U.S.

Silicon Valley Democrat to target retaliation against tech vendors amid Anthropic fallout

Democrat Rep. Sam Liccardo (Calif.) is ramping up efforts in Congress to push back on Trump administration's decision to cut the use of Anthropic's technology in federal agencies after the company's negotiations with the Pentagon over safety requirements fell apart last week. Liccardo, who represents part of Silicon Valley, announced Monday that he will introduce an amendment to the Defense Production Act this week to prohibit federal agencies from "retaliating" against high-risk technology vendors and developers that try to limit the deployment of their technology "in ways to mitigate the risk to United States citizens." The lawmaker said he will introduce the amendment during Wednesday's bill markup on the House Committee on Financial Services. "AI governance will have massive impacts on Americans, and on our future. For those experts who understand the enormity of what lies at stake, mere assertions from the Pentagon that will 'follow the law' offer little assurance where Congress has not meaningfully legislated in this space," the lawmaker, who chairs the New Democrat Coalition's Technology & Innovation Workforce Task Force, said in a statement. It comes just days after President Trump ordered federal agencies to stop using Anthropic's technology after the AI company failed to reach an agreement with the Pentagon over AI safeguards. The Defense Department (DOD) wanted language to permit the use of Anthropic's technology for "all lawful purposes, while Anthropic pushed for limitations on mass domestic surveillance and fully autonomous weapons. Following the expiration deadline last Friday, Defense Secretary Pete Hegseth also announced the Pentagon would label Anthropic as a supply chain risk in an unprecedented move for the AI industry in government. Without naming Anthropic directly, Liccardo argued such "supply chain risk monitors" and product mandates should not be used by agencies against "responsible companies" seeking guardrails. "Instead, Congress and federal agencies should learn from leading industry thinkers to better manage AI deployment," Liccardo said. The move by the Trump administration quickly spurred debate online over how much control private technology companies have over the federal government's use of its products. Ahead of negotiations falling apart Friday, Senate defense leaders from both sides of the aisle reportedly urged the DOD and technology company to resolve the dispute. Anthropic, which provided its AI models to U.S. defense and civilian agencies since late 2024, called the Pentagon supply chain risk designation "legally unsound" and risks setting a "dangerous precedent." The company said it plans to challenge the designation in court.