The US Senate is currently discussing a new legislative proposal aimed at regulating artificial intelligence companies. Under the draft, developers would be required to prove that they have implemented reasonable precautions to prevent their tools from causing harm. This duty of care is a central component of the ongoing negotiations between lawmakers.
The proposed legislation suggests granting the US Secretary of Commerce the authority to demand evidence of these safety measures. Furthermore, the Secretary could authorize government auditors to directly test AI products. Lawmakers are also considering a mechanism that would allow the federal government to block the release of AI models deemed unsafe, though companies would retain the right to challenge such decisions in federal court.
The draft bill also includes provisions that would prevent individual states from enforcing their own local regulations regarding specific risks associated with AI models. Despite these discussions, the proposal is still in the early stages of negotiation. Sources indicate that even if the bill passes through Congress, there is no guarantee it will ultimately be signed into law.
