Source: The Fulcrum

The Fulcrum analyzes the growing legislative efforts across US states to regulate AI discrimination, often called algorithmic bias. The article highlights Colorado’s Artificial Intelligence Act (SB205), effective February 2026, as the first comprehensive state-level framework targeting high-risk AI systems that influence consequential decisions in employment, education, finance, housing, and healthcare. The law mandates developers and deployers to manage discrimination risks, conduct impact assessments, and ensure transparency. However, implementation challenges have led to recommendations for refining definitions, expanding small business exemptions, and providing cure periods before enforcement.

Other states like Illinois, New York, Vermont, Texas, and Hawaii have introduced similar but varied legislation, including private rights of action and stricter penalties. The article discusses the ‘Brussels Effect’—US states adopting EU-style AI regulations—raising concerns about compliance costs versus evidence of widespread harm. The piece concludes that balancing consumer protection with innovation remains a key policy challenge, especially without federal support under the current administration.