Senate Bill Seeks to Require Federal Adoption of NIST AI Framework

A bipartisan Senate bill is seeking to require all federal agencies to adopt the National Institute of Standards and Technology’s AI Risk Management Framework.

Sens. Jerry Moran, R-Kan., and Mark Warner, D-Va., proposed the Federal AI Risk Management Act of 2023 to reduce safety and privacy risks related to the use of artificial intelligence. Rep. Ted Lieu, D-Calif., is expected to introduce companion legislation in the House of Representatives.

According to NIST’s website, the framework released in January was designed to improve the ability of all organizations to ensure the design, development, use and evaluation of AI products, services and systems can be trusted. The framework, however, is only intended for voluntary use, Moran’s office said.

Moran and Warner acknowledged that AI adoption comes with advantages, including improving operation efficiency and effectiveness and accelerating innovation, but they noted that the technology’s use should comply with NIST’s guidelines to ensure that its applications will not harm the American people.

According to Moran, organizations are facing challenges in verifying AI-generated data. Warner said AI systems should be subject to extensive testing and evaluation to ensure they operate properly.