US Senate negotiators weigh a duty-of-care requirement for frontier AI developers
U.S. Senate negotiators are discussing legislation that could require advanced AI developers to take responsibility for mitigating known major risks and could give the federal government power to block unsafe model releases, according to Reuters.
The proposal under discussion would create a duty of care for developers of advanced AI systems, requiring them to design against major known risks such as enabling nuclear or biological weapons. Reuters reported that negotiators are also considering federal authority to stop the release of an unsafe model, subject to court challenges. The legislation is still being negotiated and its final text and timing are uncertain.
A federal duty-of-care standard would move frontier AI safety from voluntary company policies toward a legal obligation. If enacted, it could materially change pre-release testing, documentation, risk controls, and launch decisions for companies building the most capable models.
Maybe — This is proposed legislation rather than a final rule, but frontier-model teams should track it because it could create concrete pre-release safety and documentation obligations.