Bipartisan AI Bill Introduced to Regulate 'Critical-Impact' AI Systems

The proposed legislation aims to provide transparency and accountability for AI tools, with a focus on systems that have significant societal impact.

Sens. John Thune (R-S.D.) and Amy Klobuchar (D-Minn.) have introduced the bipartisan Artificial Intelligence Research, Innovation, and Accountability Act of 2023, which aims to provide transparency and accountability for AI tools.
The bill would require federal agencies to create standards for AI systems and would define terms related to generative AI, including 'critical-impact' and 'high-impact' AI systems.
'Critical-impact' AI systems, under the bill, would include those used to make decisions that significantly impact the collection of biometric data without consent, the management and operation of critical infrastructure, and criminal justice.
The bill would also require 'critical-impact' AI organizations to self-certify as meeting compliance standards and would task the Commerce Department with submitting a five-year plan for testing and certifying 'critical-impact' AI.
The National Institute of Standards and Technology (NIST) would be directed to develop standards for the authenticity of online content and recommendations for technical, risk-based guardrails on 'high-impact' AI systems.