Hegseth Challenges Anthropic Over Military AI Access

Defense Secretary Pete Hegseth has issued a Friday deadline to Anthropic, demanding broader access to the company’s artificial intelligence technology or risking the loss of its Pentagon contract. The move escalates tensions between the administration and the AI firm, centered on concerns about the employ of AI in military applications, particularly those involving lethal force and domestic surveillance.

The Standoff

Hegseth summoned Anthropic CEO Dario Amodei to Washington on Tuesday for a direct meeting. According to a source familiar with the talks, Hegseth threatened to remove Anthropic from the Pentagon’s supply chain or invoke the Defense Production Act (DPA) – a Cold War-era measure granting the president control over domestic industry for national defense purposes – if the company doesn’t comply. Politico and the Associated Press both reported on the ultimatum.

Anthropic acknowledged the meeting, stating they engaged in “good-faith conversations about our usage policy to ensure Anthropic can continue to support the government’s national security mission in line with what our models can reliably and responsibly do.” However, failure to reach an agreement by Friday could result in Anthropic’s removal from the Pentagon’s supply chain and potential legal action.

Core Disagreements

The dispute stems from Anthropic’s reluctance to grant the military unrestricted access to its AI models for classified use, including applications related to domestic surveillance and autonomous weapons systems. Anthropic has voiced concerns about the reliability of current AI technology for high-stakes, life-or-death decisions, particularly in scenarios lacking human oversight. They have also advocated for new regulations governing the use of AI in mass domestic surveillance, even where legally permissible. PBS reported on these concerns.

The Pentagon, however, seeks broad access to AI technology to enhance its capabilities and counter emerging threats. Defense Secretary Hegseth emphasized in a recent memo that “AI-enabled warfare…will redefine the character of military affairs over the next decade,” and that the U.S. Military “must build on its lead” in AI development.

Potential Ramifications

Cutting Anthropic from the defense supply chain would be a significant step, typically reserved for companies linked to foreign adversaries. It would signal that the administration views the company’s continued participation as a threat. Alternatively, invoking the DPA would suggest Anthropic’s technology is considered critical to Pentagon operations, allowing the military to utilize its tools without a formal agreement. The DPA has been previously used by both the Trump and Biden administrations to address supply shortages, including medical supplies during the COVID-19 pandemic and critical minerals. The Financial Times highlighted these potential actions.

Anthropic’s Claude model is currently the only AI working on classified missions, in partnership with Palantir. The loss of this capability would necessitate the Pentagon to seek alternative AI solutions from companies like Google, OpenAI, and xAI, with whom Hegseth is reportedly already in negotiations.

The situation also raises broader questions about the balance between national security and responsible AI development, and the role of government regulation in a rapidly evolving technological landscape.