US AI hiring laws, state by state: the patchwork employers face
For multistate employers using AI in hiring, verified at source. About 5 minutes.
There is no single US law on AI in hiring. Instead there is a patchwork: New York City is live and being enforced today, Illinois and California obligations arrived in 2025 and 2026, Texas added a narrower rule, and Colorado's landmark law was delayed and then rewritten before it ever took effect. Underneath all of it, existing federal anti-discrimination law still applies. The calendar keeps moving, so the practical job is to know which rules touch where you actually hire.
The one that is live now: New York City
New York City's Local Law 144 is the rule with teeth on the ground today. Since enforcement began on 5 July 2023, an employer or employment agency may not use an automated employment decision tool to screen NYC candidates or employees unless the tool has had an independent bias audit within the past year, a summary of that audit is posted publicly, and the people evaluated were notified at least ten business days in advance (NYC DCWP, Local Law 144). Penalties run up to 500 dollars on the first day and 500 to 1,500 dollars for each day after, with each day of non-compliant use counted separately. If you hire into a New York City role, this one is not optional and not in the future.
The 2026 wave, and why the dates keep moving
Several states have now reached AI in employment. The duties differ, and so do the dates. Here is where the main ones stand as we write.
| Jurisdiction | What it reaches | Where it stands |
|---|---|---|
| New York City | Automated tools used to screen for hiring or promotion. | Live and enforced since 5 July 2023. |
| Illinois | AI that analyzes video interviews, and AI used in employment decisions under the Human Rights Act. | Video interview rule in force since 2020; the Human Rights Act amendment since 1 January 2026. |
| California | Automated decision systems used in employment, under the Fair Employment and Housing Act. | Regulations in force since 1 October 2025. |
| Texas | AI used with intent to unlawfully discriminate, plus stricter limits on government use. | In force since 1 January 2026, but narrower than the others. |
| Colorado | Developers and deployers of automated decision technology in consequential decisions, including employment. | Delayed, then rewritten; the new framework is set for 1 January 2027. |
Illinois runs on two tracks. A 2020 law governs AI that analyzes recorded video interviews, requiring notice, consent, and deletion on request (820 ILCS 42). Separately, an amendment to the Illinois Human Rights Act took effect on 1 January 2026, treating discriminatory use of AI in employment decisions as a civil rights violation and barring the use of a ZIP code as a proxy for a protected class.
California brought AI into employment through its existing discrimination law. Civil Rights Council regulations in force since 1 October 2025 make it unlawful to use an automated decision system in a way that discriminates on a protected basis, and require employers to keep related records for several years (California Civil Rights Department). They apply to employers covered by the Fair Employment and Housing Act, generally those with five or more employees, so read the threshold for your own headcount rather than assuming it covers everyone.
Texas is the one most likely to be misread. Its Responsible Artificial Intelligence Governance Act took effect on 1 January 2026 (HB 149), but for private employers it is narrower than the others: it targets AI deployed with the intent to unlawfully discriminate, and a disparate impact alone does not establish a violation. Its tougher restrictions on AI in decisions are aimed at government use. Do not treat it as a New York style bias-audit duty for private hiring.
Colorado is the cautionary tale that proves the wider point. Its 2024 Artificial Intelligence Act was first set for 1 February 2026, then pushed to 30 June 2026, and then repealed and reenacted by a new law on automated decision making technology, signed on 14 May 2026, with the revised framework set to take effect on 1 January 2027 (Colorado General Assembly). A program built to the original text would have been built to a rule that no longer exists. The lesson is not the specific date; it is that any single date in this area is provisional.
There is still no federal AI hiring law
Nothing at the federal level fills the gap. There is no AI specific hiring statute, and in 2025 the Equal Employment Opportunity Commission withdrew the AI guidance documents it had published earlier. That removal changed the guidance, not the law: Title VII, the Americans with Disabilities Act, and the Age Discrimination in Employment Act still apply to AI hiring tools, and an employer remains responsible for discriminatory outcomes a tool produces. Waiting for Washington to harmonize this is not a compliance plan.
What to do with a patchwork
- Map where you actually hire. The rules follow the role and the candidate, not your headquarters. List the cities and states where you recruit, then match each to the table above.
- Inventory your AI hiring tools. You cannot apply any of these rules until you know which tools score, rank, or screen people, and at which step.
- Handle the live ones first. New York City is enforced today, and California and Illinois are in force now. Those are real obligations, not planning items.
- Track the dates as provisional. Colorado shows why. Re-check any future date before you rely on it, and build the evidence once so it serves whichever rule lands.
Size it for your own stack
Before any of this becomes a project, the first question is which of your tools are even in scope and where. That is free and takes a couple of minutes.
Take the free, two-minute self-assessment. A few questions tell you whether your hiring tool is likely in scope and what the first step is. No email needed to see your result.
Take the free 2-minute checkWant the audit, notice, and evidence steps as editable files your team fills in? See the Toolkit (US$ 99, one-time).
Track this as it hardens. One brief a week on AI hiring rules, EU and US: each claim traced to the primary source, what is settled, what is still open, what to do next.
Free. No spam. Unsubscribe anytime.
Working through New York City specifically? Start with whether you are covered by Local Law 144, then the bias-audit checklist.
This is educational material and a starting point, not legal advice. It surveys several US jurisdictions, each with its own rules, thresholds, and changes over time, and a number of these laws are moving as we write. For how any of them applies to your specific tools, roles, and locations, consult a qualified employment lawyer in the relevant jurisdiction. Signato is not a law firm and does not certify compliance.
Every claim here is traced to the primary regulation and checked by a person before it goes out. We tell you what is settled, what is still open, and what to do next. How we work.