Pete Hegseth Seeks Military Access to Anthropic AI Models
▼ Summary
– US Defense Secretary Pete Hegseth has threatened to cut AI company Anthropic from the Pentagon’s supply chain unless it agrees by Friday to allow its technology in all lawful military applications.
– The conflict stems from Anthropic’s refusal to grant unrestricted access to its AI models for classified military use, including domestic surveillance and fully autonomous lethal missions.
– During a tense meeting, Hegseth threatened to invoke the Cold War-era Defense Production Act, which would compel Anthropic to work with the Pentagon regardless of its consent.
– A senior Pentagon official stated that failure to comply would also result in Anthropic being labeled a “supply chain risk” to the Department of Defense.
– Anthropic responded that it is engaged in good-faith discussions to support national security within the reliable and responsible limits of its models.
Defense Secretary Pete Hegseth has issued a stark ultimatum to artificial intelligence company Anthropic, demanding it agree to allow its technology to be used in all lawful military applications or face removal from the Pentagon’s supply chain. This confrontation centers on the military’s desire for broader access to advanced AI models, including for sensitive operations, while the company has expressed reservations about certain uses. The deadline for compliance is set for the end of the week, marking a significant escalation in the ongoing dispute between the defense establishment and a leading AI developer.
During a high-stakes meeting in Washington, Hegseth presented Anthropic CEO Dario Amodei with a clear choice. According to sources familiar with the discussion, the Secretary threatened to cut the company off from defense contracts or to invoke the Defense Production Act. This powerful law, originating from the Cold War, grants the president authority to direct domestic industrial production for national defense purposes. A senior Pentagon official stated that Anthropic has until Friday afternoon “to get on board or not” with the stipulated terms.
The official further warned that failure to comply would lead Hegseth to ensure the Act is invoked against Anthropic, compelling the company to provide its AI models to the Pentagon irrespective of its own corporate policies. Additionally, the Defense Department would formally designate Anthropic as “a supply chain risk,” a label that could severely impact its current and future government business. The official sought to downplay concerns about the intended applications, asserting, “You can’t lead tactical ops by exception,” and claiming the issue is unrelated to “mass surveillance and autonomous weapons.”
For its part, Anthropic has stated it remains engaged in constructive dialogue. The company emphasized its commitment to “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.” This statement suggests the firm is seeking a compromise that aligns with its internal ethical guidelines while attempting to accommodate national security requirements. The outcome of this standoff could set a critical precedent for how the U.S. government leverages, and potentially mandates, access to cutting-edge private sector AI technology for defense purposes.
(Source: Ars Technica)
