AI Moratorium Stays in US Senate Budget Reconciliation Bill

A measure that would block all state laws regulating artificial intelligence for a decade remains in the Senate version of President Donald Trump’s so-called “big beautiful bill.”

See Also: OnDemand | Fireside Chat: Staying Secure and Compliant Alongside AI Innovation

Lawmakers confirmed Sunday that the provision remains in the budget bill, even as other controversial elements such as cuts to benefits for underserved communities like the Supplemental Nutrition Assistance Program were removed. The measure only requires a simple majority vote to be cut from the final version and has sparked bipartisan opposition.

Senate rules allow expedited consideration for budget resolutions known a reconciliation legislation – provided the bill language only affects spending or revenue. Reconciliation bills in the Senate must past a hurdle known as the “Byrd Bath” – named after Sen. Robert C. Byrd, D-W.Va – in which an advisor with the title of Parliamentarian of the United States scrubs provisions that fall short of the rule.

Sen. Ted Cruz, the Texas Republican who’s the force behind the measure, told Punchbowl he tweaked the measure to get ahead of the Parliamentarian by tying the regulation moratorium to access to a $500 million federal investment fund for deploying AI.

“States that voluntarily seek these funds must agree to temporarily pause AI regulations,” Cruz’s office said.

Axios reported Monday that the reconciliation language allows the federal government to pause grants made under the $42 billion Broadband Equity, Access and Deployment grant program. The Trump administration could “rescind BEAD grants and then only restart them if states agree to the AI pause,” Doug Calidas, Americans for Responsible Innovation senior VP of government affairs, told Axios.

Trump administration AI and cryptocurrency czar David Sacks supports the moratorium, writing that he opposes “a patchwork of 50 different regulatory regimes driven by the AI Doomerism.”

Lawmakers from both parties have argued the measure threatens oversight, undermines state rights and weakens accountability for private sector technology firms. Cruz’s Democratic colleagues on the Senate Commerce Committee, including Ranking Member Maria Cantwell, Wash., and Sen. Ed Markey, Mass., have come out against it.

Tennessee Republican Marsha Blackburn opposes it, reported The Tennessean. Her state, home to the estate of rock legend Elvis Presley, approved in 2024 a law against unauthorized uses of artists’ voices and likenesses. Other Republicans to voice opposition include Missouri’s Josh Hawley and Wisconsin’s Ron Johnson.

“I personally don’t think we should be setting a federal standard right now and prohibiting the states from doing what we should be doing in a federated republic. Let the states experiment,” Johnson told The Hill earlier this month.

Parts of the private sector have also voice opposition. An association of insurance agents in a June 16 letter to Senate leaders objected to broad language, writing that it would extend the moratorium to regulations on a “wide range of processes using existing analytical tools and software that insurers rely on every day.” The National Association of Insurance Commissioners also wrote that a moratorium “could disrupt the well-established processes that state regulators use to review models and ensure fairness and transparency in insurance markets and ensure that premiums charged to the consumer are not inadequate or excessive.”

Many statehouses have continued to push AI regulation legislation. New York Senate lawmakers approved a June 11 bill that would allow the state to sue makers of frontier models that fail to publish safety and security protocols and risk evaluation or to disclose serious incidents.

Utah in March enacted the Artificial Intelligence Policy Act making users of AI models potentially liable if their chatbot violates state consumer protection law. A chatbot’s statements should be viewed no differently than an employee’s, the law holds.

A Pew Center survey of 5,410 U.S. adults published in April found nearly six out of 10 Americans are more concerned that the U.S. government won’t do enough to regulate AI. Only roughly two out of 10 Americans said their concern is that the government will regulate too much.