New Hampshire’s attorney general has joined his peers to tell Congress that they shouldn’t block state efforts to do what the federal government won’t do in terms of “protecting consumers from the harmful effects” of artificial intelligence.

Attorney General John Formella was one of 40 state attorneys general from both political parties who signed a letter voicing opposition to a sweeping amendment in the budget reconciliation bill that says states can’t enforce any laws or regulations on artificial intelligence and automated decision-making systems for 10 years.

“Imposing a broad moratorium on all state action while Congress fails to act in this area is irresponsible and deprives consumers of reasonable protections,” the letter says. 

“State AGs have stepped in to protect their citizens from a myriad of privacy and social media harms after witnessing, over a period of years, the fallout caused by tech companies’ implementation of new technology coupled with a woefully inadequate federal response. In the face of Congressional inaction on the emergence of real-world harms raised by the use of A.I., states are likely to be the forum for addressing such issues.” 

Technologies known collectively as artificial intelligence are being tested in a wide array of industries to see whether they can do everything from simple office paperwork to discovering new drug chemistries. Although the basics of machine learning have been used for years, recent advances mean the technology is improving at breakneck speed which has regulators scrambling to limit damage without throttling benefits.

Many states have proposed regulation of A.I., some looking to have strong guardrails while others have been more hands-off.

Many companies that sell or use A.I. are concerned that a patchwork of regulation will develop with each state having separate requirements or limits. But so far the federal government has shown little interest in developing a national regulatory framework.

The proposed moratorium on state regulation is part of the budget reconciliation process and was inserted by the House Energy and Commerce Committee. It has not yet been taken up by the U.S. Senate.

“State attorneys general have led the way in protecting consumers from the harmful effects of unregulated A.I. This proposed amendment would undermine those efforts, stripping away important consumer protections without offering any solution to fill the void. I stand with my colleagues in urging Congress to reject this dangerous provision,” said Formella said in a statement Monday.

“A.I. promises to revolutionize America’s economy, spur achievement and innovation, and improve lives across the country. However, the rise of such technology presents real, immediate dangers ranging from explicit material and election interference to deception, exploitation, and harassment against consumers. … In the years ahead, additional matters – many unforeseeable today given the rapidly evolving nature of this technology – are likely to arise,” he said. 

“Any effort to prohibit states from enacting and enforcing laws aimed at regulating A.I. and protecting consumers will leave A.I. entirely unregulated at any level and Americans completely exposed to its known harms and evolving, real-world risks, ultimately leading to dangerous consequences for the American people.”

The attorneys general of all six New England states signed the letter.