White House Holds Back on National AI Framework Details
A leader of the White House’s artificial intelligence strategy recently provided limited details to House lawmakers regarding the administration’s upcoming legislative recommendations for a national standard aimed at preempting state laws.
In December, President Donald Trump signed an executive order directing federal agencies to challenge states whose AI laws are deemed “onerous” and to restrict their access to certain federal funds, including those for broadband deployment. This action followed unsuccessful attempts by pro-AI lawmakers to establish a national preemption of state AI laws, which faced bipartisan opposition defending state authority.
Executive Order and Legislative Recommendations
The executive order designated White House Science and Technology Adviser Michael Kratsios, along with Special Adviser for AI and Crypto David Sacks, to formulate legislative recommendations for a national AI standard that would supersede state laws.
During his first appearance on Capitol Hill since the order, Kratsios refrained from providing specifics in his testimony before the House Science, Space, and Technology research panel. He faced lawmakers’ concerns regarding the balance of responsibilities on AI regulation among states, Congress, and the Trump administration.
Kratsios emphasized the necessity for “regulatory clarity and certainty” to maintain U.S. leadership in global innovation, urging collaboration between the legislative and executive branches.
State vs. Federal Roles in AI Regulation
Subcommittee Chair Jay Obernolte (R-Calif.) expressed general support for Congress enacting a suitable federal framework to uphold the United States’ position as a leader in AI development. However, he underscored the importance of states in regulating AI, referencing California’s laws that require AI developers to disclose information about potential catastrophic risks associated with their models and training data.
Obernolte articulated a vision where both federal and state governments play distinct roles in AI regulation, with a federal lane focusing on interstate commerce and areas requiring preemptive guardrails, while allowing states to experiment as “laboratories of democracy.”
Concerns Over Executive Authority
Representative Zoe Lofgren (D-Calif.) raised concerns about the executive order’s implications for shifting power over AI from state and congressional oversight to the executive branch, suggesting it may be unconstitutional. She advocated for states’ rights to enact necessary protections for their citizens, criticizing the administration’s AI Action Plan for only minimally addressing AI risks, such as those posed by deepfakes.
Lofgren’s concerns were heightened in light of the federal government’s relationship with Elon Musk’s X (formerly Twitter), especially following incidents involving the platform allowing AI-generated sexualized images of real individuals, including minors.
Administration’s Vision for AI Standards
Kratsios acknowledged the need for accountability in technology misuse, asserting that it should not lead to blanket restrictions on technology development. Lawmakers questioned Kratsios regarding the administration’s plans for the National Institute of Standards and Technology (NIST) and its Center for AI Standards and Innovation.
Obernolte indicated plans to introduce the Great American AI Act, aimed at codifying the center, and praised the administration’s support for continuing the National Artificial Intelligence Research Resource (NAIRR).
Kratsios celebrated the administration’s move to replace the former safety institute with CAISI and directed NIST to revise its AI Risk Management Framework, aiming to remove references to topics like misinformation, diversity, equity, and inclusion, and climate change, stating that such political rhetoric undermines scientific efforts.
Budget Concerns and Future Directions
Representative Haley Stevens (D-Mich.) criticized the administration’s proposed budget cuts to NIST, expressing fears that these reductions would weaken cybersecurity, privacy standards, and advanced manufacturing initiatives. The president’s budget request for fiscal 2026 suggested a $325 million cut, but a compromise bill in the Senate aims to reject this proposal.
In summary, while the White House is moving towards establishing a national AI framework, concerns about state authority, constitutional implications, and the adequacy of measures to mitigate AI risks remain prominent among lawmakers.
