CSOAI - United States (federal)

US federal AI policy

The US has no comprehensive federal AI law. Federal policy is set through executive action and is currently deregulatory and pro-innovation, leaving most binding rules to the states.

Approach: Executive-led, deregulatory (no omnibus federal AI law)

Key points

+One AI-specific federal statute so far: the TAKE IT DOWN Act (non-consensual AI imagery)
+Deregulatory executive orders + policy frameworks (2025-2026)
+Sector regulators (e.g. FTC) apply existing law to AI
+Binding AI rules largely sit at the state level (CO, TX, CA, UT)

Questions, answered

Does the US have a federal AI law?

No comprehensive one. Congress has enacted the TAKE IT DOWN Act (targeting non-consensual AI-generated imagery); broader federal AI policy is set by executive orders and frameworks and is currently deregulatory. Most binding AI obligations are at the state level.

How is the US federal approach different from the EU AI Act?

The EU AI Act is a binding, risk-tiered omnibus law. The US federal approach is executive-led and pro-innovation, with the substantive obligations mostly coming from individual states (Colorado, Texas, California, Utah).

Which US states regulate AI?

Notably Colorado (SB 26-189 / ADMT, which in May 2026 repealed and replaced the original SB 24-205 AI Act; effective 1 Jan 2027), Texas (TRAIGA), California (AI transparency laws) and Utah, plus NYC Local Law 144 for hiring - see the US state pages.

AI regulation worldwide is evolving fast. This reflects the position as of June 2026 and is not legal advice - verify current law for your jurisdiction.

Ask the Council assistant — US federal AI policy — obligations, who must act, and the first move
Governed answer · AI governance & cybersecurity only · signed to Layer 0
Open the full AI OS →