a day ago
Congressional Budget Bill Could Override State AI Laws
Employers increasingly face a patchwork of state laws governing the use of artificial intelligence (AI) and automated decision systems (ADS) in hiring, promotion, and workforce management. But that may change dramatically. On May 22, the U.S. House of Representatives passed the 'One Big Beautiful Bill Act' (OBBBA), a sprawling budget reconciliation package that includes a controversial provision: a 10-year federal moratorium on state and local regulation of AI.
Buried in Section 43201, the measure would prohibit any state or municipality from enforcing laws that regulate AI models, AI systems, or automated decision systems for a decade after the bill's enactment. The scope is broad. If enacted, the provision would preempt a wide range of existing state and local laws and halt momentum behind dozens of AI-related bills pending across legislatures nationwide.
Section 43201 explicitly bars state and local governments from enforcing laws that limit, restrict, or otherwise regulate AI systems, models, or automated decision systems. The proposed preemption period lasts ten years. The language is sweeping. If enacted, it would suspend enforcement of laws specifically targeting AI, including those governing how employers deploy algorithmic tools in employment decisions.
The bill defines 'artificial intelligence systems' as machine-based systems that, for a set of human-defined objectives, make predictions, recommendations, or decisions that influence real or virtual environments. 'Automated decision systems' are defined even more broadly. These include any computational process using machine learning, statistical modeling, or data analytics that outputs a classification, score, or recommendation intended to materially influence or replace human decision-making.
Taken together, these definitions likely cover tools that employers use every day, from resume screening software and video interview scoring platforms to employee monitoring systems powered by machine learning.
If passed, the moratorium would put existing state and municipal AI regulations on hold, including:
The proposal would also preempt emerging AI legislation, including numerous AI-related bills introduced across statehouses in 2025.
While the moratorium is expansive, Section 43201 does include limited exceptions. A state or local law would not be preempted if it:
These exceptions are narrow. While they may preserve state laws that apply broadly to both AI and non-AI tools performing similar functions, most current AI employment laws are unlikely to qualify. These laws often impose requirements, such as bias audits, transparency disclosures, or data handling rules, specifically tailored to AI systems, making them vulnerable to preemption under the bill. In practice, unless a state law imposes requirements that apply equally to both AI and traditional systems performing similar tasks, it is likely to be preempted.
Section 43201 was advanced through the House Energy and Commerce Committee in mid-May and included in the broader budget reconciliation bill that passed the House by a 215–214 vote. Because the bill is proceeding under budget reconciliation, it is immune to filibuster in the Senate and could pass with a simple majority.
Still, the provision faces significant obstacles. Critics argue that the moratorium is regulatory in nature and may not meet the budgetary requirements of the reconciliation process under the Byrd Rule. The Byrd Rule limits reconciliation bills to provisions that directly affect federal spending or revenue, and allows regulatory provisions, like this moratorium, to be struck before a Senate vote. If ruled out of order, the Senate Parliamentarian could remove the AI moratorium provision altogether.
Additionally, a bipartisan coalition of 40 state attorneys general has voiced strong opposition to the measure, citing concerns over states' constitutional authority to protect consumers and workers. Legal scholars have raised potential challenges under the Tenth Amendment, particularly given the moratorium's interference with state police powers related to public health, safety, and civil rights.
Even if enacted, the moratorium is likely to face immediate litigation.
For employers who use AI in employment decisions, this bill creates both uncertainty and opportunity. Until the Senate acts, state laws remain in effect. Employers should continue to assess compliance obligations under local statutes like Local Law 144 in New York City and forthcoming measures in Illinois and Colorado.
At the same time, organizations should begin scenario planning for a world in which those state-level obligations are preempted. That may include:
Even if a moratorium is enacted, employers remain responsible for the outcomes of their employment decisions. AI does not shield organizations from liability. Human oversight, fairness, and transparency should remain cornerstones of responsible hiring.
Section 43201 represents one of the most aggressive attempts yet to consolidate AI regulation at the federal level. For employers navigating a fractured and fast-evolving regulatory landscape, the measure promises temporary clarity, but at the cost of state-level experimentation and innovation.
Whether the moratorium survives reconciliation and legal challenge remains to be seen. In the meantime, employers should not assume that compliance obligations will disappear. Instead, they should continue investing in practices that promote transparency, accountability, and fairness in AI-driven employment decisions, and prepare for the possibility of a federally preempted landscape.
