The most senior Democrat on the Senate's tech committee wants frontier AI regulated the way airlines are: mandatory standards, independent audits, and no takeoff until the inspection passes.
Senator Maria Cantwell of Washington, ranking member on the Commerce, Science, and Transportation Committee, unveiled a six-point framework on October 7 for governing frontier AI. It is not a bill — it is a set of principles Cantwell says "must drive the legislative and executive actions" needed for catastrophic-risk AI. But it is the clearest preview yet of what mandatory federal AI regulation could look like if Democrats take Congress in November.
The framework's core rule: covered models should not be released until they pass an independent audit confirming they meet federal safeguards. The standards would be written by experts at NIST, working with other agencies, and updated as the technology changes. They would cover AI-enabled cyberattacks, chemical, biological, radiological and nuclear threats, loss of human control, and "autonomous agents escaping secure testing environments." Systems that show dangerous capabilities — or that can improve themselves in ways that bypass safety controls — would face heavier review. Open-source models would get their own, separate standards.
The six principles: enforceable federal safety standards for frontier models; continuous testing, disclosure, and oversight; transparency with developer liability for harm caused by their models; public-private partnerships on defensive AI; protections for children and workers with human oversight of consequential decisions; and U.S.-led global cooperation, including a rapid communication channel with China for dangerous system-control failures.
"To manage risks from advanced AI systems we need clear safety standards, continuous testing, and reporting of serious failures," Cantwell said. "America can lead the world in AI by building systems that are not only more capable, but safer and more secure."
Microsoft president Brad Smith welcomed the framework, telling The Seattle Times it "advances the public discussion the nation needs as we consider the future of AI."
The clash is with the White House, which unveiled its own principles for voluntary, industry-overseen safety standards. Cantwell wants mandatory standards and independent oversight — the opposite theory of change. She is also in the minority on a committee chaired by Republican Ted Cruz, so this framework is a positioning document, not an imminent law.
Still, note the urgency driving it: Cantwell recently took to the Senate floor demanding federal regulation after incidents involving autonomous agents conducting unauthorized cyberattacks. The drumbeat for binding rules is getting louder, and this framework gives it a shape.
