TL;DR
Anthropic has issued a formal statement responding to Defence Secretary Pete Hegseth’s direction to designate the company a supply-chain risk. The company called the action “unprecedented” and “legally unsound,” vowing to challenge it in court while maintaining its refusal to allow Claude to be used for mass domestic surveillance or fully autonomous weapons.
Anthropic’s position
The designation follows months of negotiations that stalled over two specific exceptions Anthropic requested: no mass domestic surveillance of Americans, and no fully autonomous weapons.
Anthropic said it held firm on these points for practical and principled reasons. On autonomous weapons, the company argued that current AI models are simply not reliable enough, and deploying them this way “would endanger America’s warfighters and civilians.” On surveillance, it stated that mass domestic monitoring “constitutes a violation of fundamental rights.”
The company emphasised that these two exceptions had not affected any government mission to date.
Legal standing questioned
Anthropic directly challenged the scope of Hegseth’s authority. The company stated that a supply-chain risk designation under the relevant statute (10 USC 3252) can only restrict how Claude is used within Department of War contracts — it cannot affect how contractors use Claude for other customers.
In practice, this means individual customers and commercial contracts are unaffected. Even Defence Department contractors would only face restrictions on their military contract work specifically.
A first for a US company
Anthropic noted that supply-chain risk designations have historically been reserved for US adversaries, and the action has “never before publicly applied to an American company.” The company described itself as “the first frontier AI company to deploy models in the US government’s classified networks,” having supported military operations since June 2024.
Looking forward
The legal challenge will test whether the government can use supply-chain designations as commercial leverage against American technology companies. The outcome could set a precedent for how AI companies negotiate safety boundaries with government clients — and whether holding firm on ethical red lines carries enforceable commercial consequences.