Executive Order 14365 directs federal officials to challenge state artificial intelligence laws that conflict with a national policy of a minimally burdensome framework for AI. The order was signed Dec. 11, 2025, and published in the Federal Register on Dec. 16, 2025.
The order states that United States policy is to sustain and enhance the country’s global AI dominance through that framework. It says a national standard should forbid conflicting state laws and should also protect children, stop censorship, respect copyrights, and safeguard communities. Until such a standard exists, the order says, the administration must check the most onerous and excessive state laws.
Within 30 days of the order, the Attorney General shall establish an AI Litigation Task Force. The order says the task force’s sole responsibility is to challenge state AI laws inconsistent with the stated policy. Grounds listed in the order include arguments that the laws reach across state lines in ways the Constitution does not allow, that existing federal rules already override them, or that they are unlawful for other reasons the Attorney General decides. The task force is to confer periodically with the Special Advisor for AI and Crypto, the Assistant to the President for Science and Technology, the Assistant to the President for Economic Policy, and the Counsel to the President.
Within 90 days, the Secretary of Commerce shall publish an evaluation of existing state AI laws, after consulting those same advisers. The order says the evaluation must identify onerous laws that conflict with the policy and laws that should be referred to the task force. At a minimum, it shall identify laws that make AI models change outputs that would otherwise be accurate, or that could force developers or deployers to disclose or report information in a way that would violate the First Amendment or another part of the Constitution. The evaluation may also identify state laws that promote AI innovation consistent with the policy.
The order links part of that review to broadband money. Within 90 days, the Commerce Secretary, through the Assistant Secretary of Commerce for Communications and Information, shall issue a Policy Notice on conditions for remaining funding under the Broadband Equity Access and Deployment Program. States with onerous AI laws identified in the evaluation are ineligible for non-deployment funds, as far as federal law allows. The notice must also explain how a split set of state AI rules could put at risk deployments funded by the program, AI applications that rely on high-speed networks, and the program’s mission of universal high-speed connectivity. The order says the funding was saved through administration reforms it calls Benefit of the Bargain. It does not state a dollar amount.
Executive departments and agencies are to review grant programs they control, after consulting the Special Advisor for AI and Crypto. They are to determine whether they may condition those grants on a state not enacting an AI law that conflicts with the order, including a law identified by Commerce or challenged by the task force. For a state that has already enacted such a law, the condition may instead be a binding agreement not to enforce it while the state receives the discretionary funding.
Within 90 days after the Commerce evaluation is published, the chairman of the Federal Communications Commission shall start a proceeding on whether to set a national rule on what AI models must report and disclose, one that would override state laws that conflict with it. The chairman is to consult the Special Advisor for AI and Crypto.
Within 90 days of the order, the chairman of the Federal Trade Commission shall issue a policy statement on how the Federal Trade Commission Act’s ban on unfair and deceptive acts or practices applies to AI models. The order says the statement must explain when state laws that require changes to the truthful outputs of AI models are preempted by that ban.
The Special Advisor for AI and Crypto and the Assistant to the President for Science and Technology are to write together a proposal for Congress for a uniform federal policy framework that preempts conflicting state AI laws. The order says the recommendation shall not propose preempting otherwise lawful state AI laws on child safety; computing capacity and data-center facilities, except for permitting changes that apply generally; state government procurement and use of AI; and other topics still to be determined. The order sets no deadline for that recommendation.
In its purpose section, the order cites a Colorado law banning “algorithmic discrimination” as an example. It says that law may push AI models to give inaccurate answers so they do not show a “differential treatment or impact” on protected groups. The order also refers to Executive Order 14179 of Jan. 23, 2025. This order does not revoke or amend that earlier order.
The order says it shall be carried out consistent with applicable law and only if appropriations are available. It states that it creates no enforceable right or benefit against the United States or its officers. The Department of Commerce is to pay the costs of publishing the order.