Anthropic draft federal AI safety bill

Politico reported on October 16 that Anthropic representatives had pitched Capitol Hill lawmakers on a confidential frontier-AI safety bill. The outlet said it obtained the draft and cited three people familiar with the effort. Anthropic’s spokesperson linked the proposal to its Advanced AI Framework.

The reported draft ties model suspension to foreseeable, imminent catastrophic risks and a Commerce Department request to a federal judge. It also provides for independent evaluations, transparency and incident reporting, whistleblower protections and preemption of some state requirements. It is a company-backed proposal described in original reporting, not a law already in force.