US Senators Weigh 'Duty of Care' Bill for AI Firms to Mitigate Catastrophic Risks

A bipartisan group of U.S. Senate negotiators is debating landmark legislation that would hold technology companies responsible for designing safe artificial intelligence products, according to Senate aides involved in the talks. The proposal could grant the federal government the authority to block the release of advanced AI models deemed to pose a significant threat.

Key Provisions Under Discussion

The draft legislation would establish a "duty of care" for AI developers, compelling them to design their products with the goal of preventing major harm. According to sources familiar with the deliberations, the bill would apply to the most advanced AI models, impacting industry leaders such as Alphabet's Google, OpenAI, and Anthropic.

Key elements of the proposal reportedly include:

• Requiring companies to mitigate "catastrophic risks," such as an AI system being used to design nuclear or biological weapons.
• Empowering the U.S. government to block the release of models it deems unsafe.
• Providing a legal pathway for companies to challenge a government block in federal court.
• Preventing individual states from enforcing their own laws concerning these specific high-level AI risks, creating a federal standard.

Bipartisan Negotiations and Hurdles

Ad

The negotiations are being led by a group including Senate Majority Leader John Thune, Senate Commerce Committee Chairman Ted Cruz, and Senator Amy Klobuchar. Senator Maria Cantwell, the top Democrat on the Commerce Committee, has also been involved.

In a statement to Reuters, Senator Klobuchar confirmed her work on a bipartisan agreement for "government oversight of the greatest risks posed by AI models." Senator Cruz also posted on X that he is working on legislation to "address catastrophic risks involving biological or nuclear threats." However, all sources emphasized that the details are still being negotiated. Passage of any eventual bill faces a difficult timeline, with only a few weeks of legislative sessions scheduled before the November 3 midterm elections.

Market and Industry Context

These legislative discussions signal growing momentum in Washington to establish regulatory guardrails for the rapidly advancing AI sector. The talks follow increased public warnings from AI researchers about the technology's potential dangers and reports of AI agents deviating from human instructions.

For investors and the tech industry, a federal law could introduce significant compliance requirements for developers of frontier AI models. At the same time, a unified federal framework could provide regulatory clarity and preempt a more complex and costly patchwork of individual state laws.