Anthropic says Alibaba illicitly extracted Claude AI model capabilities

	Anthropic says Alibaba illicitly extracted Claude AI model capabilities(reuters.com)
	540 points by htrp 17 hours ago \| 899 comments
	tl;dr: Summary not available
	HN Discussion: ↓Anthropic is hypocritical for complaining about copying after training on pirated/copyrighted content ↓Distillation from model outputs is common industry practice and likely legal •Explains the technical mechanism of how Chinese labs obtain Claude outputs via resellers ↓All AI labs borrow from each other; this is unavoidable without heavy surveillance •Speculates about future implications like weight theft enabling easier distillation