Senator Marsha Blackburn (R-Tenn.) and Senate Commerce Chair Ted Cruz (R-Texas) have struck a deal on a hot-button provision in President Donald Trump’s sweeping tax legislation that seeks to bar individual U.S. states from regulating artificial intelligence.
The agreement, which softens the original language of the bill, reflects the latest flashpoint in a growing national tug-of-war over who should control the future of AI governance: the federal government or the states.
Under the revised text, states would be temporarily prohibited from implementing new AI-specific regulations for five years if they wish to access a $500 million federal fund earmarked for AI infrastructure and deployment. That’s a reduction from the 10-year moratorium initially proposed in the original bill—a version Blackburn had publicly opposed.
The updated provision now exempts state laws that regulate unfair or deceptive practices, child sexual abuse material, children’s online safety, and publicity rights—a major concession to lawmakers like Blackburn who have championed legislation to protect children and consumers from the potential harms of advanced digital technologies.
The debate around state AI regulation came in the absence of comprehensive federal AI legislation, prompting states like California, Illinois, and New York to take the lead, introducing laws and proposals to govern how artificial intelligence can be used in hiring, surveillance, consumer data profiling, education, and law enforcement.
Illinois, for example, passed the Biometric Information Privacy Act (BIPA) in 2008, which has since become a national template for regulating the use of facial recognition and biometric data. California, through its California Consumer Privacy Act (CCPA) and subsequent laws, has introduced strict obligations around automated decision-making and algorithmic transparency. Other states have followed suit with proposals targeting AI use in schools, healthcare, and financial services.
These efforts have drawn increasing opposition from tech industry lobbyists and lawmakers aligned with national security and commerce interests, who argue that a patchwork of state laws creates regulatory chaos, stifling innovation and investment in AI technology. Supporters of federal preemption believe that only Congress can provide the uniformity needed to keep pace with the rapid development of AI.
But experts warn that preempting state authority risks leaving consumers unprotected. Many states see themselves as laboratories of democracy capable of moving faster than a gridlocked Congress to address evolving AI risks—especially in areas like discrimination, misinformation, or the safety of children online.
A longtime critic of Big Tech, Senator Blackburn has staked much of her political capital on pushing laws aimed at protecting minors and reining in social media giants. She is the co-sponsor of the Kids Online Safety Act (KOSA), a bill that would require platforms to take greater responsibility for shielding children from harmful content and intrusive algorithmic profiling.
Her decision to back a five-year AI regulation freeze, after opposing the original 10-year moratorium, hinges on the exemptions negotiated with Cruz.
“To ensure we do not decimate the progress states like Tennessee have made… I am pleased Chairman Cruz has agreed to update the AI provision to exempt state laws that protect kids, creators, and other vulnerable individuals,” Blackburn said in a statement Sunday.
She added that the agreement will help ensure Congress moves forward on other digital protections.
“I look forward to working with him in the coming months to hold Big Tech accountable—including by passing the Kids Online Safety Act and an online privacy framework that gives consumers more power over their data,” she said.
Blackburn’s support gives the bill fresh momentum as the Senate prepares for key votes this week. The compromise could also ease tensions within the Republican caucus, where several lawmakers—such as Sens. Josh Hawley (R-Mo.), Ron Johnson (R-Wis.), and Rep. Marjorie Taylor Greene (R-Ga.)—have criticized the federal AI provision, arguing that it infringes on states’ rights and undermines local control.
Trump’s Push for Centralized AI Policy
The AI provision is part of the broader “One Big Beautiful Bill,” President Trump’s centerpiece economic and tax reform package, which aims to consolidate and accelerate federal investments in critical technologies, energy infrastructure, and tax relief. The administration has made it clear that it views AI as a strategic priority, both economically and geopolitically, with Trump personally pushing to pass the bill before July 4.
Supporters believe that the federal moratorium is necessary to allow time for a national AI framework to take shape, avoiding a scenario where fragmented state policies hinder the deployment of technologies critical to defense, healthcare, and infrastructure. Trump has pitched the bill as part of his “America First” digital agenda, aimed at countering China’s AI dominance and securing U.S. leadership in the technology race.
Last week, the provision cleared a key hurdle when Senate Parliamentarian Elizabeth MacDonough ruled that it complies with the Byrd Rule, meaning it can remain in the budget reconciliation package. That decision allows Republicans to move forward without Democratic support, assuming the GOP can rally enough votes from its own ranks.
What’s Next?
Whether Blackburn and Cruz’s compromise will be enough to satisfy holdouts like Hawley and Johnson remains uncertain. However, many say even a five-year pause risks chilling important state-level protections, especially as AI systems continue to evolve without clear federal oversight.
The Senate is scheduled to begin voting on the package Monday morning. The clock is ticking, with Trump’s self-imposed July 4 deadline fast approaching.
If passed, the bill would mark the first major federal intervention in how AI is governed in the United States—a dramatic escalation in a battle that has so far played out largely at the state level. For supporters, it’s a long-overdue step toward national coordination. For critics, it’s a premature power grab that leaves the public vulnerable.
Either way, the deal between Blackburn and Cruz now places AI governance squarely at the heart of a broader legislative effort that could reshape the United States’ AI landscape.