The US Senate voted by a large margin to axe a proposed 10-year ban on state regulation for artificial intelligence, in a defeat for major tech firms that had backed the plan.

Lobbyists funded by Amazon, Google, Microsoft, Meta and OpenAI, along with White House AI lead David Sacks, had argued that the AI regulation ban would remove inconsistent regional rules that could hold back US innovation and allow China to take the lead in AI.

Senators disagreed, voting 99 to 1 in favour of an amendment to a major tax and spending bill that removed the wording restricting states’ AI regulation.

Social protections

The AI provision had been included in the House of Representatives version of the bill, but had for weeks attracted widespread bipartisan criticism from governors and state officials.

Originally framed as a blanket ban on AI regulation by states for 10 years, lawmakers later tied it to federal funding so that only states that rolled back AI regulations would be eligible for subsidies for rural broadband internet or AI infrastructure.

A last-minute Republican compromise effort to save the bill would have reduced the time-period to 5 years, as well as exempting certain state AI laws that aimed to protect children or a Tennessee law known as the Elvis Act that prohibits AI being used to simulate musical artists’ voices.

Senator Marsha Blackburn of Tennessee finally teamed up with Senator Maria Cantwell of Washington to introduce the amendment that struck the entire AI proposal from the bill.

Opponents had pointed out that there are as yet no meaningful federal regulations covering AI or data protection.

The vote was welcomed by civil liberties campaigners.

‘Power grab’

“The overwhelming rejection of this Big Tech power grab underscores the massive bipartisan opposition to letting AI companies run amok,” said Max Tegmark, an MIT professor and president of the Future of Life Institute, which promotes AI regulation.

“These corporations have admitted they cannot control the very systems they’re building, and yet they demand immunity from any meaningful oversight,” Tegmark said.

Blackburn said state protections should not be removed until federal legislation is in place.

New York state legislators have passed a bill that requires large AI companies to publish safety and security reports that has yet to be signed by the governor.

Last September California governor Gavin Newsom vetoed a bill that would have required safety testing and a kill switch for AI models, following lobbying by Silicon Valley companies.