Guide
US AI laws by state
In the absence of a comprehensive federal statute, US artificial-intelligence regulation has developed as a patchwork of state and city laws, each addressing different aspects of AI.
The step most authors miss
Doing CRediT right? Don’t stop at the statement.
A CRediT statement credits you inside one paper. The recognition CRediT was built for happens when those roles are tied to you, persistently. Sign in with your ORCID — free — and claim your CRediT contributions on casrai.org, the home of the standard. They become a verified, portable part of your identity, not a line that disappears into one PDF.
Free: claim your contributions, then export a journal-ready CRediT statement, schema.org structured data, JATS XML, CSV or BibTeX — and preview your public profile. A membership publishes that profile publicly and verifies the journals you serve.
Why the United States has a patchwork
Unlike the European Union, the United States has not enacted a single comprehensive federal law governing artificial intelligence. Federal activity has largely taken the form of sector-specific rules, agency guidance and voluntary frameworks such as the NIST AI Risk Management Framework. Into that gap, individual states and cities have legislated, each addressing the concerns most salient to them. The result is often described as a patchwork: a growing collection of overlapping but non-uniform measures with different definitions, triggers and effective dates, rather than one harmonised national standard.
Colorado — the first comprehensive state law
Colorado’s AI Act (Senate Bill 24-205), enacted in 2024, is widely regarded as the first comprehensive US state AI law. It targets algorithmic discrimination by developers and deployers of high-risk AI systems used in consequential decisions, such as those affecting employment, housing, lending, healthcare and other significant services, through a duty of reasonable care. Its effective date, originally set for 2026, has been the subject of amendment. The law has been influential as an early template for discrimination-focused, risk-oriented regulation at state level.
New York City and Illinois — employment and biometrics
New York City’s Local Law 144, with enforcement from July 2023, regulates automated employment decision tools, requiring independent bias audits, public disclosure of results and notice to candidates. Illinois has addressed AI in hiring through its Artificial Intelligence Video Interview Act, which concerns the use of AI to analyse video interviews and requires notice and consent, and the state’s Biometric Information Privacy Act (BIPA) governs the collection of biometric identifiers more broadly. These measures show how some jurisdictions target specific high-stakes uses rather than AI in general.
Texas, Utah and California
Texas’s Responsible Artificial Intelligence Governance Act (TRAIGA, HB 149), effective January 2026, focuses on prohibited uses and government use of AI. Utah enacted an Artificial Intelligence Policy Act addressing disclosure when consumers interact with generative AI, particularly in regulated occupations. California has advanced multiple AI-related measures, including transparency and disclosure requirements around generative AI and automated systems, building on its broader privacy laws. Each state frames its rules differently, which is why the obligations are read against each jurisdiction’s own enacted text.
Reading the patchwork
Because these laws differ in scope, definitions and timing, the patchwork is best understood by what each measure addresses rather than as a single set of rules. Some, like Colorado’s, are comprehensive and discrimination-focused; others, like NYC’s LL144, are narrow and sector-specific; still others, like Utah’s, centre on disclosure. Voluntary federal frameworks such as the NIST AI RMF and the management-system standard ISO/IEC 42001 provide common reference points that organisations use across jurisdictions, even though they are not themselves binding laws.
Key facts
At a glance
- Definition: The patchwork of US state and city AI laws, absent a federal statute.
- Colorado: SB24-205, first comprehensive state AI law; effective date amended.
- New York City: Local Law 144, bias audits for hiring tools (from July 2023).
- Texas: TRAIGA / HB 149, prohibited uses + government AI (effective Jan 2026).
- Illinois: AI Video Interview Act + BIPA biometric privacy law.
- Federal reference: Voluntary NIST AI RMF; no comprehensive federal AI law.
Common questions
FAQ
Is there a comprehensive federal AI law in the United States?+
No comprehensive federal AI statute equivalent to the EU AI Act exists. US federal activity has centred on sector-specific rules, agency guidance and voluntary frameworks such as the NIST AI Risk Management Framework, which is why states and cities have legislated, producing a patchwork.
Which US state was first to pass a comprehensive AI law?+
Colorado is widely regarded as the first US state to enact a comprehensive AI law, through Senate Bill 24-205 in 2024. It addresses algorithmic discrimination by developers and deployers of high-risk AI systems in consequential decisions, with an effective date that has been amended.
Do these state AI laws all regulate the same things?+
No. They address different concerns. Colorado focuses on algorithmic discrimination, NYC on bias audits for hiring tools, Texas on prohibited uses and government AI, and Utah on generative-AI disclosure. The definitions, triggers and effective dates differ across jurisdictions.
Going deeper
