Anthropic's leaked AI bill draws fire from safety advocates

Anthropic has spent months pitching an AI safety framework to lawmakers, and the leaked draft is that framework translated into legislative language, according to a company spokesperson. The leak pulled the proposal into public view before Anthropic could adjust it, and a top lobbyist responded by posting about tweaks the company plans to make — changes meant to better reflect the pace of AI progress and its position on pre-empting state laws. That position is now the flashpoint: AI safety advocates argue the draft doesn't go nearly far enough, unfairly pre-empts state laws, and accepts too much possible risk. For a company long seen as the safety-promoting frontier lab, the backlash has put it on the defensive, and the episode shows how a framework designed to shape federal rules can draw criticism from the very advocates such a lab might expect to count on.