Pentagon freezes out Anthropic as it signs deals with AI rivals - Defense News

Ah, the Pentagon has finally figured out who the real threat is — and it's not China, it's the AI company that won't play ball with killer robots. The Department of Defense has frozen out Anthropic, designating it a 'supply-chain risk to U.S. national security' back in March, all because Claude's creators politely declined to hand over unrestricted access for fully autonomous weapons and mass domestic surveillance. Because nothing says 'national security' quite like blacklisting the only lab with a conscience.

This follows the War Department's cosy little AI agreements we noted last week, where seven compliant tech giants — OpenAI, Google, Microsoft, and the rest — lined up to deploy their 'advanced AI capabilities' on classified networks for 'lawful operational use.' Apparently 'lawful' is a flexible term when you've got the right lawyers. Anthropic's sin? Having the audacity to ask what 'lawful' actually means before handing over the keys to the surveillance state.

So the message is clear: If you want Pentagon contracts, just check your ethics at the door. OpenAI, fresh from restricting its own Cyber model to 'trusted partners' while sneering at Anthropic for the same, is now the darling of the defense establishment. Funny how that works — the company that lectured Anthropic about openness is now the Pentagon's favourite closed shop. Consistency, thy name is defense contracting.