News for hackers. Stuff that matters.
Re:Claude Opus 4.7 by stavros on Thursday April 16, 2026 @02:07PM (#47797259)
Anthropic's issue was only that the AI isn't yet good enough to tell who's an American, so it avoids killing them. They were fine with the "killing non-Americans" bit.
Re:Claude Opus 4.7 by senderista on Friday April 17, 2026 @12:54PM (#47807963)
I don't think that is true at all, could you provide a source?
Re:Claude Opus 4.7 by stavros on Friday April 17, 2026 @01:11PM (#47808185)
https://www.anthropic.com/news/statement-department-of-war

> But today, frontier AI systems are simply not reliable enough to power fully autonomous weapons. We will not knowingly provide a product that puts America’s warfighters and civilians at risk.

Re:Claude Opus 4.7 by senderista on Friday April 17, 2026 @07:08PM (#47811550)
That’s interesting: I recall reading that sentence but I parsed it as “America’s warfighters” AND “civilians” (i.e., civilians in general), rather than as “America’s warfighters and America’s civilians”. I think the former reading is more plausible since American civilians are not normally at physical risk from American military operations. It’s unfortunate that this sentence is worded ambiguously, though.
Re:Claude Opus 4.7 by stavros on Friday April 17, 2026 @07:10PM (#47811572)
Oh interesting, I didn't parse it as "civilians in general". Hm, yes, it is unfortunate that we won't know what he meant to write.