> Simpletons read that as: "The DOW wants to kill people with AI."
replyThe DoW actually used AI in the Iran war, which led to them attacking a school and killing more than a hundred childrem. That's exactly what Anthropic wanted them to prevent. https://www.bloomberg.com/graphics/2026-iran-school-attack/
The DoD has been using AI for engagement since the 80s. Maven, the system used in your example has been active since 2017.
replyNone of those systems pull the trigger. They assemble data for human review.
You seem to be misunderstanding the facts. Claude was already used by the US Military, under an existing contract between Anthropic and the US Department of War.
replyThe core dispute was the US Military deciding it needed to renegotiate this contract, but without Anthropic's two red lines: No fully autonomous weapons (i.e., zero humans in the loop), and no mass domestic surveillance.
well if you were Anthropic and you didn't want the government using your technology in these ways, how would you phrase it? What else would you have done?
reply