AI Governance Laws Tracker for HR

Five AI Employment Laws Take Effect Before 2027. Three Already Have.

This is the real-time record of what is changing, who it covers, and what your organization needs on file before the deadline arrives. Every entry is checked against the primary source. We are not offering legal advice.

See What's Changing Get Notified of New Requirements
 
Days Until the Next Deadline
5
Requirements Taking Effect Before Jan 1, 2027
3
Already Law, Still Widely Ignored

Tracker last verified: July 16, 2026. Sources checked against primary legislative and agency text.

The Active Tracker

What's Coming, and How Long You Have

Sorted by soonest deadline. Each entry covers what it requires, who it reaches, and where the official text lives.

5 requirements

Connecticut Employment Decisions Layoffs / WARN

Connecticut AI Responsibility and Transparency Act (CART Act)

Employers filing WARN Act layoff notices must disclose whether the layoffs are related to AI use. Connecticut's anti-discrimination statute now states that using AI is not a defense to a discrimination complaint.

 
Days Left
Effective Oct 1, 2026

HR Functions Affected

Workforce reductions and layoffs (WARN filings), and any employment decision touched by automated employment-related decision technology.

Who May Be Covered

Employees and applicants affected by AI-influenced employment decisions, and employees affected by AI-related layoffs.

Organizations That Should Prepare

Any employer required to file a WARN notice in Connecticut, and any employer using automated employment-related decision technology (AEDT) in the state.

Enforcement

Connecticut Attorney General, under the Connecticut Unfair Trade Practices Act. No private right of action.

Full obligation breakdown

Signed as Substitute Senate Bill No. 5, Public Act 26-15, on June 2, 2026. A second deadline follows on October 1, 2027, when employers must give written notice to employees and applicants before using an AEDT that makes or materially influences an employment decision. That second deadline falls outside this page's active window but is worth planning for now.

How the course helps: SimpliFocus™ Step 1 (Scan) is built for exactly this, knowing where AI already touches a termination or reduction-in-force decision before a regulator asks.
Official source: Connecticut Public Act 26-15
California Synthetic Media Vendor Governance

California AI Transparency Act (SB 942)

Covered generative AI providers must offer a free AI-detection tool and support provenance disclosures, such as watermarks and metadata, for AI-generated images, audio, and video.

 
Days Left
Effective Aug 2, 2026

HR Functions Affected

Recruiting and candidate vetting, employee relations investigations, internal communications and training content.

Who May Be Covered

Not employee-specific by design. Relevant to HR wherever synthetic media touches recruiting or an investigation, such as deepfake resumes or synthetic reference videos.

Organizations That Should Prepare

Covered providers of generative AI systems used by or serving California users. HR's exposure runs through the vendor tools it relies on.

Enforcement

California Attorney General. Additional platform and device provisions phase in through 2027 and 2028.

Full obligation breakdown

Originally scheduled for January 2026, the effective date was pushed to August 2, 2026 by AB 853. The law does not target HR software directly, but any HR tool that generates or processes synthetic audio, image, or video, including AI interview platforms, sits inside its reach.

How the course helps: TRACE™'s Traceable and Reason-giving pillars are what make synthetic-media provenance a governance asset instead of a liability in HR investigations.
Official source: California SB 942
Colorado Consequential Decisions Deployer Obligations

Colorado Automated Decision-Making Technology Law (SB 26-189)

Replaces the original Colorado AI Act, repealed before it took effect. Requires advance notice, post-decision disclosure, and consumer rights for automated decision-making technology used in consequential decisions, including employment.

 
Days Left
Effective Jan 1, 2027

HR Functions Affected

Hiring, promotion, compensation, and termination, any consequential decision about employment.

Who May Be Covered

Job applicants and employees subject to a consequential decision driven by automated decision-making technology.

Organizations That Should Prepare

Developers and deployers of ADMT used in consequential employment decisions. The Attorney General must issue implementing rules by January 1, 2027, which will further define scope.

Enforcement

Colorado Attorney General, treated as a deceptive trade practice.

Full obligation breakdown

Set the record straight: the original SB 24-205, once expected to take effect June 30, 2026 with a full duty-of-care and impact-assessment framework, was repealed. Governor Polis signed the replacement, SB 26-189, on May 14, 2026. The new law narrows the obligation to disclosure and transparency, dropping the original duty-of-care, risk-management-program, and Attorney General reporting requirements.

How the course helps: This is the deployer-obligation pattern SimpliFocus™ was built to operationalize: notice, disclosure, and a defensible paper trail before the Attorney General comes asking.
Official source: Colorado SB 26-189
California Significant Decisions Risk Assessment

California ADMT Regulations (CCPA)

Employers using automated decision-making technology for significant employment decisions without meaningful human involvement must complete risk assessments, provide pre-use notices, and honor opt-out and access rights.

 
Days Left
Effective Jan 1, 2027

HR Functions Affected

Hiring, work assignment, compensation, promotion, demotion, and termination.

Who May Be Covered

Job applicants, employees, and independent contractors subject to an ADMT-driven significant decision.

Organizations That Should Prepare

Businesses subject to the CCPA that use ADMT for significant employment decisions without meaningful human review.

Enforcement

California Privacy Protection Agency (CPPA).

Full obligation breakdown

The CPPA finalized these regulations on September 22, 2025, alongside new risk-assessment and cybersecurity-audit rules under the CCPA. Compliance is required by January 1, 2027. ADMT is defined broadly: any technology that processes personal information and uses computation to replace or substantially replace human decision-making.

How the course helps: Module 5, the SimpliFocus™ Framework, maps directly onto this: risk assessment, pre-use notice, and opt-out rights are the exact deliverables the framework produces.
Official source: California Privacy Protection Agency
New York Frontier Models Vendor Governance

New York RAISE Act

Large frontier AI model developers must publish safety frameworks, report critical safety incidents within 72 hours, and register with a new Department of Financial Services oversight office.

 
Days Left
Effective Jan 1, 2027

HR Functions Affected

Indirect. This law governs the AI vendors behind enterprise HR tools, not HR processes directly.

Who May Be Covered

Not employee-facing by design. Relevant to HR through the accountability of the AI vendors it relies on.

Organizations That Should Prepare

Large frontier AI developers training models above a 1026 FLOPs threshold. HR leaders should ask whether their core AI vendors fall under this scope.

Enforcement

New York Department of Financial Services, Office of Frontier AI Safety. Penalties up to $1M for a first violation, $3M for subsequent ones.

Full obligation breakdown

Signed by Governor Hochul on December 19, 2025, as S6953B/A6453B. The Act does not create obligations for typical HR software vendors, most of whom sit well under the frontier-model threshold, but it sets the ceiling for how the largest model builders behind enterprise AI tools will be expected to govern themselves.

How the course helps: Module 3 covers exactly this vendor-side governance question: what to ask an AI vendor before deploying its tool inside a talent process.
Official source: New York State Senate, S6953B
New York City Hiring & Promotion Bias Audit

NYC Local Law 144 (Automated Employment Decision Tools)

Employers and employment agencies using an automated employment decision tool to screen candidates or evaluate employees for promotion in New York City must have it independently bias-audited every year, publish the results, and notify candidates in advance.

 
Status
In effect since Jul 5, 2023

HR Functions Affected

Hiring and promotion screening specifically. These are the only two decision types the law covers.

Who May Be Covered

Job candidates and employees being considered for promotion for a role based in New York City.

Organizations That Should Prepare

Employers and employment agencies using an AEDT for a NYC-based role, regardless of where the company is headquartered. The law does not exempt tools built in-house; a homegrown scoring or ranking tool is covered the same as a vendor product.

Enforcement

NYC Department of Consumer and Worker Protection (DCWP). Civil penalties between $500 and $1,500 per violation, per day.

Full obligation breakdown

An AEDT is defined broadly: any computational process, including one derived from machine learning or AI, that issues a simplified output, such as a score, classification, or recommendation, that is used to substantially assist or replace human decision-making in hiring or promotion. Employers must have the tool audited by an independent auditor within one year before each use, publish a summary of the audit results and the tool's distribution date on their website, and notify candidates or employees at least ten business days before use, including how to request an alternative selection process or accommodation where one is available.

A 2025 New York State Comptroller audit found that DCWP has struggled to identify non-compliance, particularly where employers never disclose that AI is being used at all. That gap is worth naming plainly: this is the requirement organizations are most likely to already be violating without knowing it.

How the course helps: The longest-running of these laws, and the one enforcement data says employers still get wrong most often. A reasonable first stop if governance work has to start somewhere.
Official source: NYC Department of Consumer and Worker Protection
Illinois Employment Decisions Civil Rights

Illinois AI Disclosure Law (Public Act 103-0804)

Amends the Illinois Human Rights Act to prohibit AI use that causes a discriminatory effect across the full employment lifecycle, bars using zip code as a proxy for protected class, and requires employers to notify workers when AI is used in a covered decision.

 
Status
In effect since Jan 1, 2026

HR Functions Affected

Recruitment, hiring, promotion, renewal of employment, selection for training or apprenticeship, discharge, discipline, tenure, and other terms, privileges, or conditions of employment.

Who May Be Covered

Job applicants and employees across every IHRA-protected characteristic: race, color, ancestry, national origin, disability, religion, sex, sexual orientation, pregnancy, age over 40, military status, citizenship, conviction or arrest record, language, work authorization status, and, since January 1, 2025, family responsibilities and reproductive health decisions.

Organizations That Should Prepare

Any Illinois employer using AI, including generative AI, anywhere in the employment lifecycle above. The law's definition of AI is broad and deliberately aligned with Colorado's and the EU AI Act's framing.

Enforcement

Illinois Department of Human Rights and the Illinois Human Rights Commission, under standard IHRA process. Remedies can include back pay, reinstatement, emotional distress damages, and attorneys' fees.

Full obligation breakdown

Public Act 103-0804 (originally HB 3773) does two things at once. First, it prohibits employers from using AI in a way that causes a discriminatory effect against a protected class, even without intent, and separately bars using zip code as a stand-in for race or socioeconomic status in AI-driven hiring. Second, it requires employers using AI for a covered employment decision to notify affected employees and applicants that AI is being used. The statute itself leaves the mechanics of that notice, timing, format, and delivery method, to rulemaking. The Illinois Department of Human Rights is the agency tasked with adopting those rules, so the practical notice requirements are still being finalized even though the underlying law has been in effect since January 1, 2026.

This is the reason the card moved up from the "Already in Effect" list: an effective date in the past does not mean the compliance picture is settled. HR leaders should treat the notice-and-disclosure mechanics as an open item to track, not a box already checked.

How the course helps: Module 2 draws the line HR needs here: this is a civil-rights compliance obligation enforced through the Illinois Human Rights Commission, not a policy an IT team can sign off on alone.
Official source: Illinois General Assembly
Illinois Frontier Models Effective 2028

Illinois Artificial Intelligence Safety Measures Act (SB 315)

Requires the largest frontier AI developers, those generating more than $500 million in annual revenue and trained using extensive compute, to publish safety and transparency frameworks, undergo independent annual audits, and report critical safety incidents.

 
Days Left
Effective Jan 1, 2028

HR Functions Affected

Indirect. Governs the frontier AI developers behind enterprise HR tools, not HR processes directly.

Who May Be Covered

Not employee-facing by design. Relevant to HR through the accountability of the frontier AI vendors it relies on.

Organizations That Should Prepare

The largest AI developers, generally above $500 million in annual revenue and high compute thresholds. HR leaders should ask whether their core AI vendors are built on a covered frontier model.

Enforcement

Illinois Attorney General. Penalties up to $1M for a first violation, $3M for subsequent violations.

Full obligation breakdown

Signed by Governor Pritzker on July 6, 2026, this is the newest addition to the growing set of frontier-model safety laws, alongside New York's RAISE Act. Its effective date, January 1, 2028, falls after this page's usual date window, but it is significant enough, and moving fast enough, to earn a place here now rather than waiting for its own countdown to start. Covered developers, the most advanced, highest-compute model builders with more than $500 million in annual revenue, must publish a public safety and transparency framework explaining how they evaluate catastrophic risk, mitigate it, and respond to incidents. The law requires yearly independent third-party audits and mandates reporting of critical safety incidents within a short window. It also creates confidential reporting channels and whistleblower protections for employees who raise AI safety concerns.

How the course helps: The same vendor-side governance question Module 3 covers for frontier developers: ask whether your core AI vendor is built on a model this law, or New York's RAISE Act, will eventually reach.
Official source: Illinois General Assembly, SB 315

Already in Effect (2026)

If You Haven't Addressed These, You're Not Early. You're Overdue.

Three requirements took effect in 2026. This list is scoped to the effective date, not the signing date, which is why Texas sits here even though TRAIGA was signed in 2025; it did not become enforceable until January 1, 2026. If your organization is just now getting oriented to AI governance in HR, this is the baseline, not the frontier.

HR leadership team reviewing AI governance documentation together

Texas Responsible Artificial Intelligence Governance Act (TRAIGA)

Effective Jan 1, 2026

Texas's statewide AI governance law regulating developers and deployers of AI systems. Unlike Colorado or Illinois, it does not mandate employment-specific bias audits or notices, but it makes intentional discriminatory AI use a regulated, high-penalty risk, enforced by the Texas Attorney General.

Read the full summary

Signed by Governor Abbott on June 22, 2025, TRAIGA applies broadly to any company or government entity that develops or deploys an AI system operating in Texas or used by Texas residents, covering both developers and deployers. It is built around categorical prohibitions rather than a bias-audit or notice regime: no one may develop or deploy an AI system intentionally aimed at inciting self-harm or criminal activity, intended to discriminate against a protected class in violation of state or federal civil rights law, used solely to produce child sexual abuse material or certain sexually explicit content, or intended to impair a person's constitutional rights. Disparate impact alone does not establish intent under the law, which is a meaningfully lower bar than Colorado's or Illinois's discriminatory-effect standards.

For HR specifically, the relevant piece is the intentional-discrimination prohibition applied to any deployer, including an employer deploying an AI hiring or screening tool with discriminatory intent. TRAIGA also requires government entities to disclose AI use to consumers, sets healthcare AI disclosure rules, and creates a regulatory sandbox letting companies test AI systems under relaxed enforcement. Enforcement runs exclusively through the Texas Attorney General, with no private right of action. Civil penalties range from $10,000 to $200,000 per violation, or $2,000 to $40,000 per day for continuing violations, with a 60-day cure period available for curable violations before the AG can pursue full penalties.

Official source: Texas Legislature Online, HB 149 Bill Analysis

Connecticut CTDPA AI Training-Data Disclosure (Public Act 25-113)

Effective Jul 1, 2026

Privacy notices must clearly disclose whether personal data is collected, used, or sold to train large language models. Distinct from the CART Act above, this amends Connecticut's data privacy law and reaches any HR chatbot, coaching tool, or analytics platform built on an LLM.

Read the full summary

This amendment to the Connecticut Data Privacy Act (CTDPA) requires covered controllers to update their privacy notices so they clearly and conspicuously disclose whether personal data is collected, used, or sold to train a large language model or other AI system. The disclosure has to be reasonably accessible, clear, and meaningful, not buried in general boilerplate. The obligation applies broadly to any business already subject to CTDPA thresholds, not just employers using "high-risk" employment AI, and it extends to vendors and processors acting on a controller's behalf. That means HR needs to confirm any LLM-backed HR chatbot, coaching tool, or people-analytics platform is accounted for in the company's privacy notice and vendor contracts. The same amendment also expands CTDPA's sensitive-data categories, including disability and treatment status, transgender and nonbinary status, neural data, and government ID numbers, and adds new protections for minors, both relevant to HR wellness, benefits, and youth-employment programs. Enforcement sits with the Connecticut Attorney General.

Official source: Connecticut Public Act 25-113

Oregon HB 2008 Privacy Amendments

Effective Jan 1, 2026

Bans the sale of precise geolocation data and the personal data of known minors under 16. Relevant to employee-monitoring and location-tracking tools, and to youth employment or campus recruiting programs that touch minors' data.

Read the full summary

Signed by Governor Kotek on June 2, 2025, and effective January 1, 2026, this amendment to the Oregon Consumer Privacy Act (OCPA) prohibits the sale of precise geolocation data, meaning data that places a person or their device within roughly 1,750 feet of an actual location, without opt-in consent. It also prohibits processing the personal data of a consumer a controller knows, or willfully disregards, is under 16 years old, for targeted advertising, data sales, or certain profiling, without consent, and requires parental consent for children under 13. For HR, the relevance is indirect but real: employee-monitoring and location-tracking tools, such as fleet, field-service, or badge-geofencing systems that rely on third-party data processors, now sit inside this rule, and so do youth-employment, internship, or campus-recruiting programs that collect data from applicants under 16. Enforcement sits with the Oregon Department of Justice.

Official source: Oregon Legislative Information System

Track Your State Legislature

Want a Broader View of What's Moving in Your Statehouse?

This tracker covers requirements with a confirmed effective date. If you want visibility into bills still moving through committee across all fifty states, Transparency Coalition publishes weekly AI legislative updates covering activity in every U.S. statehouse.

Visit Transparency Coalition's weekly legislative updates

Global AI Laws You Shouldn't Forget About

The EU AI Act

Outside this page's U.S. date window, but the framework every state law above is increasingly modeled on.

Global Watch

EU AI Act: Employment and Workforce Provisions

The EU AI Act classifies AI used in recruitment, candidate selection, performance evaluation, task allocation, and decisions on promotion or termination as high-risk. Deployer obligations were set to apply from August 2, 2026, but a 2026 amendment has pushed that deadline for employment-related systems to December 2, 2027, for organizations with EU workforce exposure.

Read the full breakdown

What counts as high-risk. Annex III of the EU AI Act designates AI systems used for recruitment or candidate selection, including targeted job advertisements, resume filtering, and candidate evaluation, as high-risk. The same classification covers systems used to decide on promotion or termination, allocate tasks, or monitor and evaluate worker performance.

What deployers must do. Article 26 sets the obligations for deployers, meaning the organizations using the AI, not just the vendors who built it. Deployers must use the system according to its instructions, assign human oversight, ensure the data fed into it is relevant and representative, and monitor its operation. Under Article 26(7), deployers must inform workers' representatives and affected workers or candidates before deployment: that AI is being used, how it functions, and what role it plays in the decision affecting them.

The timeline, and why it just changed. The original deadline for these obligations to apply to employment-related high-risk systems was August 2, 2026. In 2026, the EU negotiated a simplification package known as the Digital Omnibus on AI. The European Parliament approved it on June 16, 2026, and the Council of the EU gave final approval on June 29, 2026. Once formally published and in force, it pushes the employment-related high-risk compliance deadline to December 2, 2027. If that formal step is not completed before August 2, 2026, the original deadline applies as written, so this is one to watch closely over the coming weeks.

Why this matters for U.S.-based HR leaders. If your organization operates in the EU, employs people there, or uses an AI vendor whose tools serve EU-based staffing or recruiting, this is the framework your U.S. state law obligations are increasingly being compared against and modeled on. Colorado's and California's deployer-notice requirements borrow directly from Article 26's logic. Module 3 of AI Governance Essentials covers Article 26 in detail.

Official source: EU Artificial Intelligence Act
Cover of When AI Breaks the Law: AI Governance for Talent Leaders, by Margaret Spence

Go Deeper

This Tracker Tells You What's Changing. The Book Tells You Why It Keeps Happening.

When AI Breaks the Law: AI Governance for Talent Leaders is the operational manual behind every law on this page: the failure patterns that put HR in the room with regulators, and the framework that keeps it out. Written by the person who managed the consequences before she wrote the book about preventing them.

Get the Book

Build Your Governance Program

Every Law Above Has a Module Built to Answer It

AI Governance Essentials for HR Executives is the six-module masterclass built for exactly the deadlines tracked on this page, not a generic AI literacy course. HR by HR, designed to be defensible when the accountability moment arrives.

  • Module 2: AI Governance for HR is not IT Governance. Why vendor contracts and policy PDFs are not a program.
  • Module 3: The EU AI Act and what it means for vendor-side governance, including frontier-model laws like New York's RAISE Act.
  • Module 5: The SimpliFocus™ AI Governance Framework, the operational system behind every deployer obligation on this page.
  • Bonus: New Laws Affecting AI in HR, updated as new requirements are issued.
Enroll in AI Governance Essentials
AI Governance Essentials for HR Executives course cover

Frequently Asked

What HR Leaders Are Actually Asking

What AI employment laws are taking effect in the United States in 2026 and 2027?
Between today and January 1, 2027, five requirements take effect: Connecticut's CART Act (October 1, 2026), California's AI Transparency Act (August 2, 2026), and three laws effective January 1, 2027, Colorado's Automated Decision-Making Technology Law, California's ADMT Regulations, and New York's RAISE Act. Illinois's AI Disclosure Law, Texas's TRAIGA, Connecticut's CTDPA amendment, and Oregon's privacy law are already in effect. New York City's Local Law 144 has applied since 2023. Illinois also just signed a second, later law, the Artificial Intelligence Safety Measures Act, tracked here even though it does not take effect until January 1, 2028.
Does my company need to comply with the Colorado AI Act?
The original Colorado AI Act was repealed before it took effect. It has been replaced by SB 26-189, effective January 1, 2027, which applies to developers and deployers of automated decision-making technology used in consequential decisions, including employment. If your organization uses AI to help decide hiring, promotion, compensation, or termination for anyone in Colorado, this applies to you.
Does Illinois have more than one AI law affecting employers?
Yes, two separate ones. The AI Disclosure Law (Public Act 103-0804), effective January 1, 2026, amends the Illinois Human Rights Act and covers discriminatory AI use and employee notice across hiring, promotion, and other employment decisions. The Artificial Intelligence Safety Measures Act (SB 315), signed July 6, 2026, is a separate frontier-model safety law aimed at the largest AI developers, not at HR decisions directly, and does not take effect until January 1, 2028.
Does Texas require AI bias audits or employee notices like Colorado and Illinois?
No. TRAIGA, effective January 1, 2026, does not mandate bias audits or AI-use notices. It instead prohibits specific intentional misuses of AI, including intentional discrimination against a protected class, and is enforced exclusively by the Texas Attorney General. Disparate impact alone does not violate the law; intent has to be shown.
What is the difference between a developer and a deployer under state AI laws?
A developer builds or substantially modifies the AI system. A deployer is the organization that uses it, which means most HR departments are deployers, not developers, even when the tool comes from a vendor. Deployer obligations typically include notice, disclosure, and risk assessment. Developer obligations typically include documentation and testing support. Most state laws place the heavier operational burden on deployers.
What counts as a high-risk or consequential decision AI system in employment?
Generally, any AI tool that makes or is a substantial factor in decisions about hiring, promotion, compensation, discipline, or termination. The exact definition varies by state. Colorado and California both use consequential decision language; earlier frameworks used high-risk. Read the specific statute before assuming a tool falls outside the definition.
Do AI hiring law obligations apply to small businesses, or only large employers?
It depends on the law. Some, like Illinois's amendment to the Human Rights Act, apply broadly with few size exemptions. Others, like New York's RAISE Act, only reach the largest frontier AI developers and carry no direct HR obligation. Check the covered employer definition on each entry above; do not assume small size means no exposure.
What is an AI impact assessment, and who has to complete one?
A written assessment covering the AI system's purpose, the data it uses, how it was tested, what risks it presents, and what safeguards are in place. Deployers of covered high-risk or consequential-decision AI systems are typically responsible for completing it, not the vendor who built the tool.
What happens if my organization doesn't comply with a state AI employment law?
Enforcement varies by state. Several of these laws, including Connecticut's CART Act and Colorado's ADMT law, are enforced exclusively by the state Attorney General as unfair or deceptive trade practices, with no private right of action. Others, like Illinois's amendment, allow civil penalties up to $5,000 per violation. None of them offer a formal safe harbor. Documented governance is the closest thing to one.
Is there a federal AI employment law, or is this state-by-state only?
As of this tracker's last update, there is no comprehensive federal AI employment law. Coverage is state by state, which is why an organization operating in multiple states can face several different obligations for the same AI tool.
How is the EU AI Act different from U.S. state AI employment laws?
The EU AI Act is a single comprehensive framework covering the full lifecycle of AI systems, with deployer obligations under Article 26 for employment use. U.S. state laws are narrower, more numerous, and inconsistent with one another. An organization with EU workforce exposure needs both compliance tracks, not just one.
What is NYC Local Law 144 and does it apply outside New York City?
Local Law 144 requires an annual independent bias audit of any automated employment decision tool used to screen candidates or evaluate employees for promotion in New York City, with results published and candidates notified. It applies to use of the tool for a role based in New York City, regardless of where the employer is headquartered.
Do AI-generated interview or resume-screening tools trigger disclosure requirements?
In most states with a notice requirement, yes. If the tool processes, scores, or ranks candidates, or materially influences a hiring decision, disclosure obligations typically apply, whether the interview involves synthetic video, automated scoring, or resume ranking.
What should HR do first to prepare for an upcoming AI compliance deadline?
Start with an inventory: know every place AI touches a hiring, promotion, compensation, or termination decision before documenting anything else. You cannot govern what you have not mapped. This is the first step in the SimpliFocus™ AI Governance Framework for exactly that reason.
How often should this tracker be checked for updates?
This tracker is verified monthly at minimum, and immediately after any state passes new legislation or an agency issues implementing rules. The last-verified date at the top of the page shows exactly how current the information is.
Where can I find the official legal text for these AI employment laws?
Every entry on this page links directly to a primary source, the bill text, public act, or agency regulation page, not a law firm summary. Use the Official Source link on each card.

Bring This to Your Team

Book a Speaker or Start Building Your Governance Program

Every law on this page becomes a liability question the moment your board or general counsel asks about it. Margaret works directly with HR executives and leadership teams to build the answer first, through keynotes, workshops, and enterprise AI governance engagements.

Tell us what you need. Responses within one business day.

Serving organizations across North America & globally
See the Tracker