State AI Law Is the Only AI Law. Everywhere It's Crumbling.
Summary
The 2024 Colorado AI Act, initially intended to regulate AI use, has been significantly weakened, passing as Senate Bill 189. This revised law now only mandates that companies inform individuals after an AI system has been used to deny them a loan, job, or housing, and provide an opportunity to appeal. The actual use of AI in these life-altering decisions is no longer restricted, only the disclosure of its involvement. This trend of legislative erosion is not isolated to Colorado; similar AI regulation bills in Texas, California, and New York have also been either heavily modified or vetoed, indicating a widespread difficulty in establishing comprehensive state-level AI governance.
Key takeaway
For CTOs and legal counsel evaluating AI deployment risks, the current landscape of state AI law suggests a fragmented and permissive regulatory environment. Your teams should prioritize establishing clear post-decision notification and appeal mechanisms for AI-driven outcomes, particularly in critical areas like lending, employment, and housing, rather than expecting broad restrictions on AI use itself. This shift minimizes immediate compliance burdens but highlights the need for internal ethical guidelines.
Key insights
State-level AI regulation efforts are broadly failing, resulting in minimal oversight.
Principles
- Disclosure is replacing restriction in AI law.
- AI governance faces significant legislative hurdles.
In practice
- Audit AI systems for post-decision notification.
- Prepare appeal processes for AI-driven denials.
Topics
- State AI Law
- AI Regulation
- Colorado Senate Bill 189
- Legislative Vetoes
- AI Disclosure Requirements
Best for: CTO, VP of Engineering/Data, Executive, Policy Maker, Legal Professional, Director of AI/ML
Related on AIssential
Editorial summary, takeaway, and curation by AIssential. Original article published by Artificial Intelligence.