The CREATE AI Act of 2025 would establish the National Artificial Intelligence Research Resource (NAIRR), a federal program intended to broaden access to the computing power, data, software, and testbeds needed for artificial intelligence research. The bill’s stated purpose is to reduce the concentration of AI research resources among large technology companies and make those resources available to researchers, educators, students, small businesses, and other eligible U.S.-based entities. It directs the National Science Foundation, in coordination with the Office of Science and Technology Policy and other federal partners, to stand up the NAIRR within one year of enactment.
The bill creates a governance structure that includes a NAIRR Steering Subcommittee, a Program Management Office within NSF, advisory committees, and a competitively selected nongovernmental Operating Entity. These bodies would oversee day-to-day operations, select and manage resources, publish reports, evaluate performance, and update the program over time. The NAIRR would provide a mix of computational resources, curated datasets, educational tools, and access to AI testbeds, with explicit attention to privacy, ethics, civil rights and civil liberties, safety, scientific integrity, cybersecurity, and research security.
If enacted, the bill would amend the National Artificial Intelligence Initiative Act of 2020 and related federal law to add a new Title LVI establishing the NAIRR and its governing framework. It would give NSF and OSTP new responsibilities to coordinate federal AI research infrastructure, authorize agencies to contribute resources or funding, and allow the NAIRR to accept private donations of cash, services, and personal property. The bill would also create eligibility rules for users, security and privacy review requirements, fee authority with a free access tier, and reporting obligations that would affect federal agencies, research institutions, nonprofits, small businesses, and other eligible users of AI research infrastructure.
The bill appears generally favorable and bipartisan in concept, as reflected by its introduction by Representatives Obernolte and Beyer, who are from different parties. The text emphasizes expanding access, democratizing AI research, and strengthening U.S. leadership in artificial intelligence, suggesting a broad policy consensus around improving research capacity. No committee transcript or vote record is available in the provided material, so there is no recorded floor or committee debate to indicate broader opposition or support beyond the bill’s bipartisan sponsorship and policy framing.
The main areas of potential contention are likely to be governance, access, security, and funding. The bill gives significant coordinating authority to NSF, OSTP, and a nongovernmental Operating Entity, which could raise questions about oversight and accountability. It also includes eligibility limits, excluding users employed by certain foreign countries, and requires privacy, ethics, civil rights, safety, and research-security reviews, which may draw scrutiny over implementation burdens and access restrictions. Another possible point of debate is the fee structure and reliance on appropriations and private donations, which could raise concerns about whether the program will be adequately funded and whether access will be equitable across user groups.