Senate reintroduces bipartisan bill to set federal AI standards and testing

A bipartisan group of senators reintroduced the Future of AI Innovation Act on Thursday, pushing to establish voluntary federal guidelines, standards, and testing infrastructure for artificial intelligence across industry and the public sector.

What happened

Senators Maria Cantwell, Todd Young, John Hickenlooper, and Marsha Blackburn, reintroduced the Future of AI Innovation Act. The bill would authorize the National Institute of Standards and Technology's Center for AI Standards and Innovation to develop AI guidance, standards, and performance benchmarks.

The legislation would also create a testbed program coordinated by NIST, the Department of Energy, and the National Science Foundation, leveraging DOE's national labs to evaluate the capabilities and limitations of AI systems. Additional provisions include a prize competition to incentivize innovation, a standards coalition with U.S. allies, and directions for federal science agencies to release curated public datasets covering agriculture, medicine, transportation, and manufacturing.

Industry and advocacy organizations including the Software & Information Industry Association, the Alliance for Secure AI, and Americans for Responsible Innovation have already signaled support for the bill.

The backstory

A previous version of the Future of AI Innovation Act advanced out of the Senate Commerce, Science, and Transportation Committee by voice vote in 2024, as part of a package of legislation. The bill also builds on recommendations from the National AI Advisory Committee (NAIAC), a panel of outside experts that advises the government on AI. Cantwell and Young also sponsored the legislation that created NAIAC. The committee has not held a public meeting since President Trump took office.

Going deeper

The bill's components include:

• NIST's Center for AI Standards and Innovation (formerly the AI Safety Institute) would develop voluntary guidance, standards, and performance benchmarks for AI.
• A testbed program coordinated by NIST, DOE, and NSF would evaluate AI systems' capabilities and limitations using national lab resources.
• A prize competition would incentivize AI innovation.
• A standards coalition would be established with U.S. allies.
• Federal science agencies would make curated datasets publicly available for use in AI applications.
  
  

What was said

In a statement included in the Senate Commerce Committee press release, Sen. Cantwell said, "This legislation brings together private sector and government experts to develop voluntary standards for AI, create new assessment tools, and conduct testing that will ensure the United States leads in AI-driven innovation and competitiveness for decades to come."

In the know

NIST's Center for AI Standards and Innovation is the federal body tasked with developing technical guidance and safety benchmarks for AI systems. Testbeds are controlled environments used to evaluate how technologies perform under real-world conditions. The DOE's national lab network, which includes institutions like Argonne and Oak Ridge, provides high-performance computing resources that can be used to run large-scale AI evaluations. NAIAC is an advisory body of outside experts that makes recommendations to the federal government on AI policy.

Why it matters

For organizations operating in regulated industries, including healthcare, voluntary federal AI standards carry weight. Guidelines developed through NIST often become benchmarks for compliance and procurement decisions. If passed, this bill would give NIST formal authority to develop AI performance standards, which could shape how AI tools used in clinical settings, billing, and records management are evaluated and adopted.

The bottom line

With voluntary standards, testbeds, and public datasets on the table, organizations that adopt and track NIST guidance now will be better positioned as federal benchmarks solidify.

FAQs

What does NIST do?

NIST is a federal agency that develops technical standards and guidelines used across industries.

What is the difference between a voluntary and mandatory standard?

A voluntary standard is one organizations can choose to adopt, while a mandatory standard is legally required.