Skip to main content

GUARDRAILS Act 2026 Explained: The Bipartisan Push to Save State AI Protections

When President Trump signed an executive order in December 2025 attempting to halt state-level AI regulations, he ran into an unlikely coalition. More than 50 Republican state lawmakers---a contingent not known for championing regulatory authority---sent a letter to the White House urging Washington to back off. Now Congress is responding with legislation that would formally dismantle the administration's moratorium and restore states' power to protect citizens from AI harms.

The GUARDRAILS Act, short for "Guaranteeing and Upholding Americans' Right to Decide Responsible AI Laws and Standards Act," represents Congress's most direct challenge yet to the federal AI preemption debate that has consumed statehouses for over a year. Introduced simultaneously in both chambers in late March 2026, the bill has attracted unusual bipartisan support---but faces an uphill climb in a Congress divided on nearly everything except hostility toward tech regulation.

What the GUARDRAILS Act Actually Does

The legislation targets one specific weapon in the administration's arsenal: the December 2025 executive order that threatened states with federal funding cuts if they enacted AI policies deemed "inconsistent" with federal priorities.

Specifically, the GUARDRAILS Act would:

Sen. Chris Murphy (D-Conn.), who co-sponsored the Senate version alongside Sen. Brian Schatz (D-Hawaii), framed the stakes bluntly: "Donald Trump only cares about protecting the profits of his billionaire buddies in the AI industry. States have to be able to protect American families from a technology that could destroy millions of jobs, addict kids to algorithmic feeds, and discriminate in lending and hiring."

The House companion bill counts five Democratic cosponsors led by Rep. Don Beyer (D-Va.), with Rep. April McClain Delaney (D-Md.) calling the executive order an attempt to "make the internet a more dangerous place."

How This Differs From Trump's National AI Framework

The GUARDRAILS Act arrives precisely as the administration released its own National AI Policy Framework in late March 2026---a document that doubled down on federal preemption. Where Trump's approach would invalidate state-level algorithmic discrimination standards and establish federal agencies as the sole AI cops, the GUARDRAILS Act takes the opposite approach: let states regulate, and let the federal government butt out.

The philosophical divide is significant. Trump's framework treats state AI laws as obstacles to innovation that must be cleared before industry can flourish. The GUARDRAILS Act treats those same state laws as legitimate exercises of democratic governance that Washington has no business overruling.

State AI Laws Now Under Threat

If the executive order takes effect, it would invalidate some of the most consequential consumer protections passed by states in recent years. Here are the laws most at risk:

Colorado SB24-205: The most comprehensive state AI law in the country, requiring developers and deployers of "high-risk" AI systems to conduct bias audits, complete impact assessments, and notify consumers when AI influences consequential decisions about them---hiring, lending, healthcare, housing. The law took effect February 1, 2026, with enforcement beginning June 30. Penalties reach $20,000 per violation.

Texas TRAIGA (HB 149): Signed by Gov. Greg Abbott in June 2025 and enforceable since January 1, 2026, the Texas Responsible AI Governance Act prohibits AI systems designed to manipulate behavior causing harm, discriminate based on protected characteristics, or enable government social scoring. The Texas AG can impose penalties up to $200,000 per violation.

California SB 53: The Transparency in Frontier Artificial Intelligence Act requires large AI developers (those with $500 million+ in annual revenue) to publish safety frameworks, report catastrophic risks to the state quarterly, and disclose critical safety incidents within 15 days---or 24 hours if death or serious injury is imminent. Civil penalties reach $1 million per violation.

These laws represent billions in compliance costs for tech companies and years of legislative work by consumer advocates. The GUARDRAILS Act would ensure they survive.

Political Dynamics: Why Republicans Are Backing State Rights

The GUARDRAILS Act's most striking feature isn't its policy---it's its coalition. The bill counts only Democratic sponsors in Congress. But over 50 Republican state legislators signed a letter opposing federal preemption, arguing that "state-led efforts are fully consistent with conservative principles."

This isn't ideological posturing. For many red-state lawmakers, federal preemption represents Washington overreach regardless of the industry involved. And in Texas particularly, where TRAIGA passed with bipartisan support, there's genuine appetite to keep AI regulation in Austin rather than hand it to federal agencies the state can't directly hold accountable.

Sen. Marsha Blackburn (R-Tenn.) has conditioned any Republican support for AI preemption on Congress first passing the Kids Online Safety Act---a child protection bill with bipartisan backing. If KOSA stalls, preemption becomes harder to justify to conservatives.

Industry and Advocacy Reaction

Tech companies have largely supported federal preemption, arguing that a patchwork of 50 different state AI laws creates compliance nightmares that stifle innovation. The tech lobby's preferred outcome is a single federal standard that preempts stricter state rules.

Consumer advocacy groups strongly back the GUARDRAILS Act. They argue that state attorneys general have been the most effective enforcers of consumer protection in the AI space---the same state AGs who drove the Equifax settlement and the Facebook/Cambridge Analytica accountability actions. Stripping their authority to enforce AI-specific rules hands Big Tech exactly the regulatory vacuum it wants.

Privacy advocates are more divided. Some see the GUARDRAILS Act as the only realistic path to preserving consumer protections. Others worry that state-by-state regulation creates gaps that sophisticated companies will exploit.

What Happens Next

The GUARDRAILS Act faces long odds. The Republican-controlled Congress has shown little appetite for curbing executive authority, and the White House has made its position clear: federal preemption serves American competitiveness.

But the unusual Republican state coalition changes the calculus. If enough pressure builds from statehouses, especially in Texas and other jurisdictions with AI laws already on the books, Congress may face pressure to act. The alternative---allowing the executive order to invalidate state protections while no federal standard exists---creates a regulatory vacuum that even industry backers struggle to defend.

The clock is ticking. Colorado's enforcement window opens June 30. Texas AGs are already preparing cases under TRAIGA. Without Congressional action, states face an impossible choice: enforce their own laws and risk federal funding cuts, or step back and leave citizens unprotected.

Photo by Sora Shimazaki on Pexels

Frequently Asked Questions

What exactly is the GUARDRAILS Act and who introduced it?

The GUARDRAILS Act (Guaranteeing and Upholding Americans' Right to Decide Responsible AI Laws and Standards Act) is federal legislation that would repeal President Trump's December 2025 executive order attempting to preempt state AI regulations. Senate versions were introduced by Sens. Brian Schatz (D-Hawaii) and Chris Murphy (D-Conn.) with six Democratic cosponsors. House companion legislation was introduced by Rep. Don Beyer (D-Va.) with four additional Democratic cosponsors.

How does the GUARDRAILS Act differ from Trump's AI executive order?

Trump's executive order establishes a federal moratorium on state AI policies and threatens states with federal funding cuts for enacting "inconsistent" regulations. The GUARDRAILS Act would repeal that order entirely, prohibiting the use of federal funds to implement it. Where Trump's approach centralizes AI governance in Washington, the GUARDRAILS Act restores authority to states.

What specific state AI laws would be protected under the GUARDRAILS Act?

The legislation would protect state AI laws including Colorado's SB24-205 (bias audits, impact assessments, consumer notifications), Texas's TRAIGA (prohibited AI practices, AI Advisory Council), and California's SB 53 (frontier AI safety frameworks, incident reporting). These laws govern hiring, lending, healthcare, housing, and other consequential decisions involving AI.

What are the chances of the GUARDRAILS Act passing in the current Congress?

The bill faces significant obstacles with Republicans controlling both chambers and the White House opposing it. However, over 50 Republican state legislators have opposed federal preemption, creating potential pressure on Congress. The bill's fate may depend on negotiations over the Kids Online Safety Act, which some Republicans have conditioned their preemption support on.

How do tech companies and advocacy groups view the GUARDRAILS Act?

Tech companies generally support federal preemption as creating a clearer compliance environment. Consumer advocacy groups strongly back the GUARDRAILS Act, arguing that state attorneys general have been effective AI enforcers. Privacy advocates are divided, with some seeing it as the only path to preserve consumer protections and others concerned about state-by-state regulatory gaps.

What would happen if the GUARDRAILS Act becomes law vs. if it fails?

If passed, the executive order would be voided and states could enforce their AI laws without federal interference. If it fails, the administration can implement the moratorium, potentially withholding federal funds from states with AI regulations. Colorado's enforcement begins June 30, 2026; Texas AGs are already preparing TRAIGA cases. Without Congressional action, states face pressure to choose between federal funding and consumer protection.