Dario Amodei has been unusually candid about what keeps him up at night: AI used to surveil citizens at scale, weapons that fire without human intervention, technology that stamps out “pockets of disloyalty” before they can grow. He’s been writing those worries in public since last year.
Now the Pentagon — which traditionally is supposed to be the guardrail around unacceptable use — is threatening to blacklist Anthropic as a “supply chain risk” if the company won’t lift its restrictions and allow, in their words, “all lawful use.”
I broke this down on CNN today. The core issue isn’t just one contract. It’s the same question underneath every AI story right now: who controls this technology, and under what terms?
