“Anthropic's advocacy for the high-risk nature of AI models has resulted in premature and disruptive regulatory "whiplash".”