According to The Information, the Trump administration is close to creating a national artificial intelligence regulatory system. This initiative could lead to significant changes in AI regulation not only in the US but also in the global tech sector. This system would shift AI policy development to Washington, clearing the space for the multitude of state rules developers face today.
This is important because many of the best AI technologies are developed by American companies. Rules adopted in Washington will undoubtedly impact the operations of enterprises and other authorities outside the US. However, one important question remains open: how will the US government regulate the use of open-source AI?
Washington Wants One Set of Rules Instead of Fifty
The main focus here is the White House's National Framework for Artificial Intelligence, published on March 20, 2026. While the name might suggest otherwise, the report is neither a regulation nor an executive order. Experts from Georgetown University's Center for Security and Emerging Technology (CSET) believe this framework is merely a legislative proposal for Congress and cannot be considered working policy.
According to CSET analysts Mina Narayanan, Jessica Ji, and Vikram Venkatram, this document should be seen as a starting point for negotiations. Instead of assuming Congress will adopt the proposed framework in its current form, they view it as a way for the administration to express its priorities while allowing Congress to create its own AI laws.
The core idea of this proposal is simple: eliminate numerous AI regulatory rules across different states and develop a single federal standard. The initiative is based on an executive order issued in December 2025, which tasked White House officials with developing recommendations for a nationwide AI policy.
This initiative followed a lack of progress from Congress on proposals to ban states from enacting new AI laws or to tie certain federal funding to states limiting their own AI regulatory rules. These proposals reflected concerns that companies developing AI systems would have to deal with a multitude of different regulations.
This problem is no longer hypothetical. Colorado's AI law introduced rules to prevent algorithmic discrimination in high-risk AI projects. Texas is also reported to have announced its own transparency and consumer protection rules, while California continues to work on issues related to AI legislation for frontier models and AI-generated content. Although approaches differ, they all contribute to an increasingly fragmented regulatory landscape.
The White House argues that a unified federal system would provide developers with more transparent rules when addressing issues such as child safety, AI-generated deepfakes, and the innovative opportunities offered by regulatory sandboxes. If Congress takes this step, businesses could spend less time adapting their products to each state's laws and more time on development.
Trump's AI Strategy Extends Beyond Congress
Legislation passed by Congress is just one component of the administration's plan. On June 2, 2026, President Trump signed Executive Order No. 14409 titled "Promoting Innovation and Security in Advanced Artificial Intelligence," which called on federal agencies to strengthen cybersecurity for advanced AI technologies.
The directive calls for the creation of an AI cybersecurity information sharing hub, operating jointly with the private sector, to accelerate the discovery of software vulnerabilities. Furthermore, the directive establishes the concept of "covered advanced models," referring to the best AI systems, and requires agencies to develop a plan for their safe use.
Both the legislative framework and the executive order affirm the administration's core idea: maintaining the United States' leadership in artificial intelligence while simultaneously strengthening its cybersecurity capabilities.
The Open-Source Question Remains Unresolved
The greatest uncertainty lies in what will happen to open-source AI models. On one hand, the draft discusses important topics such as child safety, digital authenticity, deepfakes, and innovation. On the other hand, it does not provide a clear definition of how open-source AI should be regulated. The Information believes this is one of the main unresolved issues of this proposal.
Debates are unfolding against a backdrop of increasing global attention on advanced AI systems. The 2026 International AI Safety Report, authored by Turing Award winner Yoshua Bengio and compiled by over 100 experts from more than 30 countries, concludes that increasingly powerful AI systems require stronger governance. The report also notes that open-source AI models pose unique governance challenges because once released, they can be freely modified and distributed, making consistent implementation of safety measures difficult.
For AI developers, the current focus might not be on open-source per se, but rather on regulatory clarity. A federal system that takes precedence over state rules could transform disparate compliance requirements into a single national standard and thus minimize costs for companies doing business in every state. Conversely, it means states will have less opportunity to develop their own AI safeguards before federal authorities enact new rules.
It remains unclear whether Congress will ultimately approve this proposal. However, even if it is just a proposal, this policy model indicates that authorities in Washington would like to become the primary body responsible for regulating artificial intelligence technologies. If American companies still produce a large portion of the world's AI technologies, it's evident that the final decision will impact not only US laws but also the rules many developers worldwide must follow.
