10 min read
Lesson 2 of 5
US AI Policy & Executive Action
From the course AI Regulation & Policy Landscape
The United States takes a different approach to AI regulation — relying more on executive action, sector-specific rules, and voluntary commitments than comprehensive legislation.
Federal Executive Orders
Key executive actions on AI:
- EO 14110 (Oct 2023) — required safety testing for powerful AI, NIST AI safety standards, and agency AI use guidelines
- OMB Memoranda — federal agency AI governance requirements including rights-impacting AI safeguards
- CHIPS and Science Act — funding for AI research and development infrastructure
- National AI Initiative — coordinating federal AI R&D across agencies
Sector-Specific Regulation
Different agencies regulate AI in their domains:
- FTC — enforcement against deceptive AI practices, algorithmic discrimination
- FDA — approval process for AI-powered medical devices (500+ cleared to date)
- SEC — AI in financial services, algorithmic trading oversight
- EEOC — AI in hiring and employment decisions
- DOT/NHTSA — autonomous vehicle safety standards
- FCC — AI in telecommunications and spectrum management
State-Level AI Laws
States are moving faster than federal legislation:
- Colorado AI Act — first comprehensive state AI law, focused on high-risk decisions
- California AI Bills — multiple proposals on deepfakes, hiring AI, and safety testing
- Illinois BIPA — biometric information privacy affecting facial recognition AI
- NYC Local Law 144 — audit requirements for AI in hiring
- Texas, Virginia, Connecticut — various AI transparency and disclosure requirements
Voluntary Industry Commitments
The White House secured commitments from major AI companies:
- Safety testing before release (red-teaming)
- Watermarking AI-generated content
- Sharing safety research and best practices
- Investing in AI safety and security research
- Developing technical standards for AI safety
Congressional Efforts
Congress is developing AI legislation in several areas:
- Deepfake regulation (DEEPFAKES Accountability Act)
- AI in elections (Protect Elections from Deceptive AI Act)
- Algorithmic accountability (Algorithmic Accountability Act)
- AI research funding (National AI Research Resource)
- National security AI (various defense-focused bills)
Key Takeaways
- •The US relies on executive action and sector-specific regulation rather than comprehensive law
- •FTC, FDA, SEC, and EEOC each regulate AI in their domains
- •States like Colorado and California are passing their own AI laws
- •Voluntary industry commitments cover safety testing and content watermarking
- •Congressional AI legislation focuses on deepfakes, elections, and accountability
0 of 5 lessons marked complete
Progress is kept in this browser only.