Skip to content

WASHINGTON – U.S. Senators Jim Risch (R-Idaho) and Ben Ray Luján (D-N.M.), with Senators Dick Durbin (D-Ill.), John Thune (R-S.D.), and Marsha Blackburn (R-Tenn.) introduced the Testing and Evaluation Systems for Trusted Artificial Intelligence (TEST) AI Act to improve the federal government’s capacity to test and evaluate Artificial Intelligence (AI) systems to drive innovation, protect national security, and safeguard Americans against risks.

“AI has the potential to revolutionize our research and innovation across a myriad of industries; however, it also attracts malevolent actors who aim to exploit its capabilities, posing substantial threats to privacy and national security,” said Risch. “The Idaho National Laboratory is already a leader in national security and cybersecurity, and the TEST AI Act will rightly use the National Lab’s capabilities to establish safeguards and systems to prevent misuse of this growing technology.”

“The use of Artificial Intelligence has reached every sector – ranging from medicine, to financial institutions, to governments and businesses. But our country cannot ignore the vulnerabilities and risks that come with it. While these systems have the power to change lives, they can also fall short by providing inaccurate or biased data, and are at risk of malicious attacks or misuse by our adversaries,” said Luján. “The TEST AI Act addresses these shortcomings by developing government testbeds for testing and evaluation of AI systems. This will help harness the talent of our National Laboratories to boost our federal capacity to implement responsible guardrails that protect our national security, ensuring this technology is not misused or manipulated.”

The TEST AI Act directs the National Institute of Standards and Technology to coordinate with the Department of Energy to establish testbeds for testing and evaluation of trusted AI systems to advance AI tools, capabilities, and workforce needs; improve reliability and trustworthiness of commercial and federal AI systems; and establish testbeds, including classified testbeds as necessary, to support safeguards and systems to test, evaluate, and prevent misuse of AI systems.

# # #