On 11 December 2025 the US President signed Executive Order 14365, Ensuring a National Policy Framework for Artificial Intelligence, published in the Federal Register on 16 December.
Two core actions
First, a “minimally burdensome” federal policy orientation. The order’s policy statement states that it is US policy to “sustain and enhance the United States’ global AI dominance” and to achieve this through a “minimally burdensome national policy framework for AI”. This framing makes regulatory burden itself a policy variable, with a deregulatory rather than tightening orientation.
Second, an AI Litigation Task Force. The order requires the Attorney General, within 30 days, to establish an AI Litigation Task Force whose responsibility is to challenge state AI laws inconsistent with the order’s policy — on grounds including that such laws unconstitutionally regulate interstate commerce, are preempted by existing federal regulations, or are otherwise unlawful in the Attorney General’s judgment.
Why litigation rather than legislation
US AI regulatory fragmentation comes from the state level: many states have enacted or are considering laws on AI model outputs, algorithmic transparency and deepfakes, producing state-by-state compliance requirements. Traditionally, unifying rules would require Congress, but legislation is slow and uncertain.
The order takes a faster but more contested route — using the federal government’s litigation capacity to challenge state laws. In substance, it hands the question of whether the federal government should set a uniform rule to the courts rather than the legislature.
It should be noted that the order cannot directly void state law; it can only assert preemption through litigation. Its effect therefore depends on court rulings, and in the short term it produces greater legal uncertainty rather than immediate uniformity.
Three implications for companies
First, the US compliance baseline is more uncertain in the short term. State laws may remain effective while challenged, or may be stayed. Companies deploying AI in the US should assess both the federal policy orientation and the specific requirements of target states, preparing separate plans for key states. See our multi-jurisdiction service.
Second, this orientation forms a coherent line with later policy. In March 2026 the White House published a national AI legislative framework to lay groundwork for a federal standard in Congress; in June 2026 a model review executive order compressed the review window. The sequencing shows a dual-track US approach: federal uniformity and reduced burden, alongside strengthened security review. See our notes on the White House national AI framework and the US AI model review executive order.
Third, Chinese AI companies entering the US should separate “regulatory burden” from “market access”. Lighter regulatory burden does not mean wider market access. For Chinese AI companies, foreign investment review, cross-border data and export controls are access-side rules on a separate track from substantive AI regulation. See our multi-jurisdiction service.
A judgment
The order reflects a distinctly American approach: under an “innovation first” goal, treating regulatory fragmentation as a cost to be removed rather than a protection to be preserved. This contrasts sharply with the EU’s risk-tiering path toward uniform rules.
For companies, it is important to keep “policy orientation” and “actual obligation” separate. The orientation may be loosening, but until courts rule, state laws can still constitute real compliance obligations.
This article is based on US Executive Order 14365, Ensuring a National Policy Framework for Artificial Intelligence, and public reporting. It is for general reference only and does not constitute legal advice.