Anthropic won’t budge as Pentagon escalates AI dispute

Anthropic faces a Friday deadline from the Pentagon to grant unrestricted access to its AI model or face punitive action, such as being declared a "supply chain risk" or having the Defense Production Act (DPA) invoked to force compliance. Anthropic maintains its refusal to allow its technology for mass surveillance or fully autonomous weapons, citing its core safety policies. Pentagon officials counter that military technology use should adhere to U.S. law, not private contractor policies. Experts warn that invoking the DPA over AI guardrails would be a significant expansion of the law's modern use and signal instability for businesses. Anthropic is reportedly not planning to ease its restrictions. The dispute is intensified by the fact that Anthropic is the only frontier AI lab with classified DOD access, leaving the Pentagon with limited redundancy, despite a memo urging agencies to avoid single-vendor dependence on classified AI systems.

(Source：TechCrunch)