Future of Artificial Intelligence Innovation Act of 2026
- Sponsor
- Sen. Young, Todd [R-IN]
- Committees
- Commerce, Science, and Transportation Committee (primary)
- Last action
- Mar 19, 2026
Bottom line
The bill aims to bolster U.S. leadership in artificial intelligence by fostering innovation, developing robust standards, enhancing research and development, and ensuring national security in AI technologies through interagency coordination and international partnerships.
What it actually does
This bill establishes a Center for Artificial Intelligence Standards and Innovation within NIST to develop voluntary best practices, metrics, and tools for AI systems, including red-teaming and synthetic content detection. It mandates an interagency program to facilitate AI testbeds, including a voluntary foundation models test program, and promotes international cooperation on AI standards. Furthermore, it directs the Office of Science and Technology Policy to prioritize federal investment in public data for AI systems and establish a program for federal grand challenges in AI research and development. The bill also includes provisions for research security and expands NIST's authority to hire critical technical experts.
Proponents argue
Proponents argue that this bill is crucial for maintaining the United States' competitive edge in the rapidly evolving field of artificial intelligence. They contend that establishing clear, voluntary standards and testbeds will accelerate responsible AI development, enhance the trustworthiness and security of AI systems, and protect against misuse by foreign adversaries. Furthermore, supporters believe that federal investment in public data and grand challenges will stimulate groundbreaking research and create a robust ecosystem for AI innovation across all sectors of the economy, benefiting companies of all sizes and strengthening national security.
Opponents contend
Opponents might argue that while the goals are laudable, the bill's reliance on voluntary standards may not be sufficient to address the complex risks posed by advanced AI systems, potentially leading to inconsistent adoption and insufficient safeguards. Some might also contend that the federal government's role in prioritizing data and grand challenges could inadvertently steer innovation in specific directions, potentially stifling organic, market-driven research or creating bureaucratic hurdles that slow down, rather than accelerate, AI development. Concerns could also be raised about the potential for significant unfunded mandates or the administrative burden on agencies tasked with implementing these broad initiatives without explicit appropriations.
The bill is moderately long and contains technical language related to artificial intelligence, requiring a dedicated read to fully grasp its implications. The table of contents helps navigate its structure.
Sec. 101, National Institute of Standards and Technology Act, Section 22B
Establishment of the Center for Artificial Intelligence Standards and Innovation (CAISI)
This provision directs the National Institute of Standards and Technology (NIST) to establish a new center, the 'Center for Artificial Intelligence Standards and Innovation' (CAISI), within 90 days of the bill's enactment. The center's mission is to help the private sector and federal agencies develop voluntary best practices for assessing AI systems. Its functions include supporting AI red-teaming and blue-teaming, developing tools for detecting synthetic content, coordinating international standards, and providing resources for federal agencies to evaluate AI systems.
Supporters argue
Proponents argue that establishing CAISI at NIST is essential for the U.S. to lead in responsible AI development. They contend that a dedicated center will provide the necessary technical expertise and coordination to develop robust, voluntary standards and testing methodologies, which are critical for ensuring the safety, security, and trustworthiness of AI systems across various applications. This will foster innovation while mitigating risks.
Critics contend
Some might argue that relying solely on voluntary standards, even with a dedicated center, may not be sufficient to address the rapid pace and potential risks of AI development. Critics could suggest that a more regulatory approach or stronger enforcement mechanisms might be necessary to ensure widespread adoption of best practices, especially concerning critical applications or national security. Concerns might also be raised about the potential for bureaucratic inefficiencies or the pace at which the center can produce relevant guidance.
Tradeoffs
The provision balances the desire for rapid AI innovation, which often thrives in less regulated environments, with the need for responsible development and risk mitigation. It seeks to achieve this through voluntary, consensus-based standards rather than prescriptive regulation, aiming to be agile but potentially less universally enforceable.
Sec. 102
Interagency Program to Facilitate Artificial Intelligence Testbeds
This section mandates the Under Secretary of Commerce for Standards and Technology (NIST) and the Secretary of Energy, in coordination with the Director of the National Science Foundation (NSF), to establish a testbed program within one year. This program will encourage collaboration between National Laboratories, Federal laboratories, NIST, NSF, and public/private entities to test, evaluate, and assess the security of AI systems. It specifically includes a voluntary test program for foundation models and prioritizes identifying security vulnerabilities in critical areas like offensive cyber capabilities and CBRN threats. The program is set to sunset after 7 years.
Sec. 111
International Coalitions on AI Standards and Innovation
This section directs the Under Secretary of Commerce for Standards and Technology and the Secretary of Energy to jointly lead international information exchange and coordination among federal agencies and with the private sector and like-minded foreign governments. The goal is to ensure effective U.S. engagement in developing and using international technical standards for AI. It requires forming alliances to support private sector-led standards, advocate for U.S. approaches, facilitate international R&D collaboration, and develop government-to-government infrastructure. Crucially, it establishes technology trust criteria for partner countries and prohibits the People's Republic of China from participating until it complies with WTO commitments.
Sec. 201, National Artificial Intelligence Initiative Act of 2020, Section 5103A
Public Data for Artificial Intelligence Systems
This section amends the National Artificial Intelligence Initiative Act of 2020 to require the Director of the Office of Science and Technology Policy (OSTP) to develop a list of priorities for federal investment in creating or improving curated, publicly available federal government data for training and evaluating AI systems. The priorities include advancing novel AI, focusing on federally funded scientific research, and datasets unlikely to receive private sector support. The Interagency Committee is tasked with establishing initiatives for these datasets, ensuring privacy, and prohibiting data sharing with countries detrimental to U.S. national security. Datasets created or improved will be made available to federal inventories and potentially to the National AI Research Resource pilot program.
Sec. 202, National Artificial Intelligence Initiative Act of 2020, Section 5107
Federal Grand Challenges in Artificial Intelligence
This section amends the National Artificial Intelligence Initiative Act of 2020 to establish a program, led by the OSTP Director, to award prizes for solving specific, well-defined, and measurable challenges in AI. The program aims to expedite AI development and stimulate R&D in areas like microelectronics, computing advancements, explainability, advanced manufacturing, and cybersecurity for AI. Eligibility for prizes is limited to U.S.-incorporated private entities or U.S. citizens/permanent residents. Various agency heads are authorized to establish prize competitions consistent with the identified priorities. The program is set to sunset after 5 years, and the GAO is directed to conduct studies on the efficacy of federal prize competitions.
Sec. 303
Certifications and Audits of Temporary Fellows
This section establishes requirements for 'temporary fellows' (contractors, consultants, etc., who are not federal employees) performing work for agencies related to artificial intelligence and other critical and emerging technologies. Before starting work, both the temporary fellow and the head of the agency must sign a certification stating that the fellow will not perform any 'inherently governmental functions.' Copies of these certifications must be submitted to the Office of Management and Budget and relevant congressional committees. Additionally, the Inspector General of each agency using temporary fellows for this Act must perform annual audits to assess compliance, including the number of fellows, their employers, work performed, and hiring authorities.
Prohibits any entity under the ownership, control, or influence of the government of a 'covered nation' (as defined in 10 U.S.C. 4872) from accessing the resources of the Center for Artificial Intelligence Standards and Innovation. The Director of NIST, with relevant agencies, must establish criteria for identification.
Sec. 101(h)
Why it matters:This is standard legislative practice for incorporating existing definitions and ensuring comprehensive coverage of prohibitions. Its placement at the end of the section is common for limitations or exclusions.
Case for: Supporters argue this is a vital safeguard to prevent foreign adversaries from exploiting U.S. AI research and development efforts, protecting national security and intellectual property. It ensures that federal resources are not inadvertently used to benefit hostile governments.
Case against: No specific opposition is typically raised against such national security provisions, though some might argue that overly broad definitions of 'covered nation' or 'influence' could inadvertently restrict legitimate international scientific collaboration or impact entities with minimal foreign ties.
Estimated impact: Significantly restricts access to federal AI standards and innovation resources for entities linked to countries like China, Russia, Iran, and North Korea, impacting their ability to leverage U.S. advancements.
Requires that any advanced computing resources, testbeds, expertise, or other resources provided by the Department of Energy (DOE) or National Laboratories to other entities under this subtitle (like NIST or NSF) must be on a reimbursable basis, pursuant to a reimbursable agreement. However, the Secretary of Energy may waive this requirement if deemed necessary or appropriate to carry out DOE missions.
Sec. 104(b) and (c)
Why it matters:This is standard administrative language to ensure proper accounting and resource management between federal agencies. The waiver provision adds flexibility, which is also a common legislative tool.
Case for: Proponents argue that requiring reimbursement ensures that DOE and National Labs are compensated for their valuable resources, preventing undue strain on their budgets and ensuring fair allocation of costs across federal initiatives. The waiver provides necessary flexibility for strategic collaborations.
Case against: Critics might argue that requiring reimbursement could create financial barriers or administrative delays for interagency collaboration, potentially slowing down critical AI research and development efforts. They might suggest that a more streamlined, non-reimbursable approach could foster faster and more integrated federal AI initiatives.
Estimated impact: Affects the financial and administrative mechanisms of interagency collaboration on AI testbeds and resources, potentially influencing the speed and scope of partnerships between DOE/National Labs and other agencies.
Mandates that the Interagency Committee, in carrying out its functions related to public data for AI systems, 'shall ensure that no data sharing is permitted with any country that the Secretary of Commerce, in consultation with the Secretary of Defense, the Secretary of State, the Secretary of Energy, and the Director of National Intelligence, determines to be engaged in conduct that is detrimental to the national security or foreign policy of the United States.'
Sec. 201(b)(4), National Artificial Intelligence Initiative Act of 2020, Section 5103A(b)(4)
Why it matters:This is a critical national security safeguard, often placed within the operational details of data governance. Its specific wording and interagency consultation requirement reflect a careful legislative intent.
Case for: Proponents argue this is a necessary and robust measure to protect sensitive U.S. data and AI research from exploitation by hostile foreign governments. It ensures that federal efforts to expand public data for AI do not inadvertently aid countries acting against U.S. interests.
Case against: No specific opposition is typically raised against such national security provisions. However, some might argue that the broad discretion given to the Secretary of Commerce, even with consultation, could lead to arbitrary or overly restrictive determinations that hinder legitimate international scientific collaboration or data exchange.
Estimated impact: Significantly restricts international data sharing for federally curated AI datasets, impacting scientific collaboration with certain countries and reinforcing U.S. national security posture in AI.
Establishes strict eligibility requirements for recipients of prize awards under the Federal Grand Challenges in AI program: private entities must be incorporated in and maintain a primary place of business in the United States, and individuals (singly or in a group) must be U.S. citizens or permanent residents.
Sec. 202(d)(2)(A), National Artificial Intelligence Initiative Act of 2020, Section 5107(d)(2)(A)
Why it matters:This is a standard legislative practice to ensure that the benefits of federally funded or incentivized programs primarily accrue to domestic entities and individuals, reflecting a 'Buy American' or 'America First' principle in research and innovation.
Case for: Proponents argue that these eligibility requirements ensure that federal investments in AI grand challenges directly benefit the U.S. economy, workforce, and national security. They contend that it prevents foreign entities from leveraging U.S. taxpayer-funded initiatives to gain a competitive advantage or access sensitive technologies.
Case against: Some critics might argue that such strict eligibility criteria could limit the pool of talent and innovation, potentially excluding highly capable international researchers or companies with significant U.S. operations but not a primary place of business. They might suggest that a more inclusive approach could foster broader scientific collaboration and accelerate global AI progress, which could ultimately benefit the U.S. as well.
Estimated impact: Ensures that the direct economic and intellectual benefits of the Federal Grand Challenges in AI program are retained within the United States, impacting the composition of participants and the ultimate beneficiaries of the prize awards.