@jd_pressman @teortaxesTex @prionsphere If it's true that Anthropic used pretty much the same constitution for Claude 2 (anthropic.com/news/claudes-c…) for Claude 3 as they implied in the technical report, the result is maybe the strongest piece of evidence for alignment by default I've ever seen
cited on: claude-2
Reproduced against link rot, credited and linked to its original. Yours and you’d rather it weren’t here? Open an issue.