NYC LL144 vs. Illinois AIVIA vs. Colorado SB24-205: AI Hiring Law Comparison

NYC LL144 vs. Illinois AIVIA vs. Colorado SB24-205: AI Hiring Law Comparison

Three jurisdictions. Three different approaches to regulating AI in hiring. Three separate compliance obligations that may all apply to the same employer.

If your company hires in New York City, Illinois, and Colorado — and uses AI tools in any part of that process — you are operating under three overlapping regulatory frameworks with different requirements, different triggers, different enforcement agencies, and different penalties.

This article compares NYC Local Law 144 (NYC Admin. Code § 20-870), Illinois AIVIA (820 ILCS 42), and Colorado SB24-205 (Colo. Rev. Stat. § 6-1-1701) side by side — explaining what each law requires, which tools trigger each law, and how a multi-state employer builds a compliance program that covers all three.

Overview Comparison: All Three Laws at a Glance

Law 1: NYC Local Law 144 — Bias Audit Focus

NYC Local Law 144 (NYC Admin. Code §§ 20-870 through 20-875, implementing rules at 6 RCNY § 5-300) is the most established of the three laws, having been in full effect since July 5, 2023. It focuses specifically on bias auditing — the statistical analysis of whether AI hiring tools produce adverse impact across demographic groups.

The Core Requirement: Annual Bias Audit

Every employer or employment agency that uses an "automated employment decision tool" (AEDT) to evaluate candidates or employees for positions located in NYC must:

1. Complete an annual independent bias audit of the AEDT before using it on candidates. The auditor must have no employment or financial relationship with the employer or AEDT vendor for at least two years prior.

2. Post the audit results publicly on the company website before the AEDT is used. The posting must include the audit date, auditor name and contact, and selection rates by demographic category (including intersectional race/sex categories).

3. Provide written notice to candidates and employees at least 10 business days before the AEDT evaluates them. The notice must identify that an AEDT is being used, what data it processes, and how to request an alternative.

What Is an AEDT Under NYC LL144?

An AEDT is any computational process — including machine learning, statistical modeling, or AI — that produces simplified output (score, rank, classification) used to substantially assist or replace discretionary employment decision-making. This broad definition captures most AI resume screeners, predictive hiring tools, automated video scoring platforms, and AI-based assessments.

Penalties and Enforcement

NYC Admin. Code § 20-875 sets penalties at $375/day for first violations and $1,500/day for subsequent violations, per AEDT, per violation type. The NYC DCWP enforces the law and has been actively conducting enforcement proceedings since 2024.

See our complete NYC LL144 Compliance Guide 2026 for the full analysis, and our NYC LL144 Enforcement article for current enforcement status.

EmployArmor tools for NYC LL144:

• Bias Audit Tool — Auditor coordination, statistical analysis, annual scheduling
• NYC LL144 Public Disclosure Page Tool — Compliant disclosure hosting
• Candidate Disclosure Notices Tool — 10-day notice generation and tracking

Law 2: Illinois AIVIA — Video Interview Consent Focus

The Illinois Artificial Intelligence Video Interview Act (AIVIA), codified at 820 ILCS 42/1 through 42/25, takes a fundamentally different approach from NYC LL144. Where LL144 focuses on bias auditing of AI tools broadly, AIVIA focuses on a specific category of tool — AI-analyzed video interviews — and centers its requirements on candidate consent and data protection.

Background: Why Illinois Chose the Video Interview Focus

Illinois enacted AIVIA in 2019, making it one of the earliest AI hiring statutes in the country — predating NYC LL144 by years. Illinois legislators focused on AI video interview analysis because companies like HireVue had begun widely deploying facial expression and speech pattern analysis tools that candidates found opaque and that raised serious civil liberties concerns.

AIVIA was subsequently amended, with the most significant expansion effective January 1, 2026, adding bias review requirements that bring it closer to NYC LL144's scope.

AIVIA's Core Requirements (as of 2026)

Consent Before Analysis. Under 820 ILCS 42/10, employers must obtain written consent from candidates before using any AI to analyze video interview recordings. The consent must explain: (1) that AI will be used, (2) the general types of characteristics the AI will assess, and (3) how the AI will be used in the evaluation process. Employers cannot use AI video analysis on candidates who have not consented.

Data Sharing Restrictions. Under 820 ILCS 42/15, employers may not share AI video interview data with any third party except for limited purposes (software operation, legal compliance). Sharing candidate video data or AI analysis outputs with other employers or data brokers is prohibited.

Data Deletion. Employers must delete candidates' video interviews within 30 days of a request for deletion. If a candidate is not hired, the employer must delete video data within 30 days of the final employment decision, unless the candidate consents to longer retention.

Bias Review (2026 Amendment). Employers using AI video interview analysis must conduct a bias review at least annually to assess whether the AI system produces different outcomes based on the candidate's race or sex. Unlike NYC LL144, AIVIA does not require the results of this review to be posted publicly — but the review must be conducted and documented.

What Tools Trigger AIVIA?

AIVIA specifically covers AI-analyzed video interviews — software that records a candidate answering questions on video and then uses AI to analyze the video content (facial expressions, speech patterns, word choice, vocal tone) to generate a score, ranking, or recommendation.

Tools that trigger AIVIA:

• HireVue (if AI scoring is enabled)
• Modern Hire (AI-analyzed video interviews)
• Spark Hire with AI analysis features
• Any other platform that records candidate video and uses AI to produce evaluative output

Tools that do NOT trigger AIVIA:

• Live video interviews (Zoom, Teams) without AI analysis
• Video interviews that are only recorded for human reviewers, without AI scoring
• Text-based or audio-only AI assessment tools

Enforcement

AIVIA is enforced by the Illinois Attorney General. The AG can investigate complaints and seek civil remedies. The law does not specify a per-day penalty structure like NYC LL144 — remedies are determined through the civil enforcement process.

Law 3: Colorado SB24-205 — Impact Assessment Focus

Colorado Senate Bill 24-205, codified at Colo. Rev. Stat. § 6-1-1701 et seq., takes the broadest approach of the three laws. Rather than focusing on specific tool types (like Illinois AIVIA) or a specific requirement (like NYC LL144's bias audit), Colorado SB24-205 establishes a comprehensive framework for "high-risk" AI systems — including those used in employment decisions — requiring pre-deployment impact assessments, ongoing monitoring, and candidate notice.

Effective Date

Colorado SB24-205 takes effect on June 1, 2026, making it the newest of the three laws. Employers should be building compliance programs now.

Who Must Comply: Developers and Deployers

Colorado's law distinguishes between AI system "developers" (companies that create AI systems) and "deployers" (companies that use AI systems in their business). Employers using AI hiring tools are "deployers" and face specific deployer obligations. AI vendors who create the tools are "developers" with separate obligations.

SB24-205 focuses on "high-risk artificial intelligence systems" — AI systems that make or substantially influence "consequential decisions" including employment decisions (hiring, firing, promotion, compensation). Most AI hiring tools that would qualify as AEDTs under NYC LL144 also qualify as high-risk AI systems under Colorado's law.

Colorado SB24-205's Core Requirements for Employers

Impact Assessment Before Deployment. Before deploying a high-risk AI system for employment decisions, employers must complete a documented impact assessment that includes: (1) an evaluation of the system's purpose and intended use, (2) an analysis of reasonably foreseeable risks of algorithmic discrimination, (3) a description of the data used and any known limitations, and (4) a description of measures taken to mitigate identified risks.

Bias Mitigation Measures. Based on the impact assessment, employers must implement reasonable measures to mitigate identified risks of algorithmic discrimination. Unlike NYC LL144, there is no mandated statistical methodology — the law requires reasonable efforts proportional to identified risks.

Candidate Notice. Employers must provide candidates with notice that a high-risk AI system is being used in employment decisions affecting them. The notice must be provided in a manner that is "reasonably likely to be seen" by the candidate. Unlike NYC LL144's strict 10-business-day requirement, Colorado's notice requirement is less precisely defined on timing.

Annual Review. Deployers must conduct an annual review of deployed high-risk AI systems to update the impact assessment and continue bias mitigation.

Transparency Reporting. Deployers must maintain documentation of impact assessments and be prepared to provide this to the Colorado AG upon request.

Penalties

Violations of Colorado SB24-205 can result in civil penalties up to $20,000 per violation. The Colorado AG and the Division of Civil Rights share enforcement authority.

Multi-State Employer Scenario: Hiring in All Three Jurisdictions

Imagine you are the HR director of a 500-person company headquartered in Chicago with offices in Denver and New York City. Your hiring process uses:

• An AI resume screening tool (ATS with ML scoring)
• A video interview platform with AI analysis
• A predictive fit scoring model at the final hiring stage

Here is what you need to comply with all three laws:

For Your NYC Hiring (LL144 Compliance):

• Commission annual independent bias audits for all three AI tools
• Post bias audit summaries (with selection rates by demographic category) publicly on your website for each tool
• Send candidates 10-business-day advance notices before each tool evaluates them
• Maintain an accommodation/alternative process for candidates who request one

Timeline: All three requirements must be in place before any NYC candidate is evaluated.

For Your Illinois Hiring (AIVIA Compliance):

• Obtain written consent from all candidates before using AI video interview analysis
• Implement data deletion workflows — delete video data within 30 days of non-hire decisions
• Restrict video data sharing to permitted purposes only
• Conduct an annual bias review of your video interview AI tool

Timeline: As of January 1, 2026, all of these are active requirements.

For Your Colorado Hiring (SB24-205 Compliance):

• Complete an impact assessment for each high-risk AI system before deployment (or for tools already deployed, complete assessments as soon as possible before June 1, 2026)
• Document bias mitigation measures for each tool
• Provide candidates with notice of AI use in consequential decisions
• Conduct annual reviews of impact assessments

Timeline: Effective June 1, 2026.

Overlap Opportunities: Where Requirements Align

Some compliance work can serve multiple frameworks:

• Bias audit documentation created for NYC LL144 can inform (but not replace) the impact assessment required by Colorado SB24-205
• Candidate notices required by NYC LL144 and Colorado SB24-205 can be combined into a single comprehensive notice
• Annual review cycles can be coordinated so audits, bias reviews, and impact assessment updates occur on the same schedule
• Data handling policies for Illinois AIVIA (deletion, no sharing) can be extended to cover Colorado's data documentation requirements

Where Requirements Differ

Despite overlap, key differences require separate compliance steps:

• Illinois AIVIA requires explicit consent before AI video analysis — you cannot simply provide notice and proceed, you must actually obtain affirmative consent
• NYC LL144 requires public posting of audit results — your impact assessment and bias review documentation under the other laws do not need to be publicly posted
• NYC LL144 has the strictest timing requirement — 10 business days — while Colorado's notice timing is less prescribed
• Colorado SB24-205 distinguishes developer vs. deployer obligations, which may affect your relationship with your AI vendors and what documentation they owe you

Which Tools Trigger Which Laws

Effective Dates and Current Enforcement Status

NYC LL144 is the most immediately pressing compliance obligation — penalties have been accruing since July 2023, and DCWP is actively enforcing. Illinois AIVIA (as amended) became enforceable in January 2026. Colorado SB24-205 enforcement begins June 1, 2026, but impact assessments should be in progress now.

Frequently Asked Questions: Multi-State AI Hiring Law Comparison

Do all three laws apply to the same employer at the same time?

Yes, if the employer hires candidates in NYC, Illinois, and Colorado using AI tools. All three laws operate independently. Compliance with one does not satisfy the others. A multi-state employer must evaluate their obligations under each law separately and build a compliance program covering all applicable requirements.

Which law is the most burdensome to comply with?

For most employers, NYC LL144 is the most complex due to its combination of three requirements (audit, disclosure, notice) and the public posting obligation. Illinois AIVIA adds the consent requirement specific to video interviews, which requires workflow changes. Colorado SB24-205 requires the most documentation through its impact assessment process. All three should be taken seriously — penalties under each can be substantial.

Is a bias audit under NYC LL144 the same as an impact assessment under Colorado SB24-205?

No, though they have similarities. NYC LL144's bias audit requires a specific statistical methodology (selection rates, adverse impact ratios, intersectional analysis) conducted by an independent auditor, with results publicly posted. Colorado SB24-205's impact assessment is a broader documented analysis of risks, mitigation measures, and data practices — less prescriptive on methodology but covering more ground. Results are not required to be posted publicly under Colorado's law. Documentation from one can inform the other, but they are separate requirements.

Does AIVIA apply to employers headquartered outside Illinois who interview Illinois candidates?

Yes. Illinois AIVIA's applicability is based on where the candidate is located and being evaluated, not where the employer is headquartered. An employer in any state that uses AI video interviews to evaluate Illinois candidates must comply with AIVIA for those interviews.

What is the relationship between NYC LL144 and the federal EEOC?

NYC LL144 is a NYC administrative law enforced by DCWP. The EEOC separately enforces Title VII (42 U.S.C. § 2000e) at the federal level. Compliance with LL144 does not insulate an employer from Title VII disparate impact claims if their AI tool produces biased outcomes. Both enforcement regimes operate simultaneously.

Does Colorado SB24-205 require publicly posting impact assessments like NYC LL144 requires public disclosure?

No. Colorado SB24-205 requires employers to maintain impact assessment documentation and make it available to the Colorado AG upon request — but there is no public posting requirement. NYC LL144's public disclosure requirement is more stringent in this respect.

Can a single consent or notice satisfy both NYC LL144 and Illinois AIVIA?

A single notice can potentially satisfy both laws if it contains all required elements of each. However, NYC LL144 requires the notice to be provided at least 10 business days before AEDT use, while AIVIA requires consent (not just notice) before AI video analysis. Notice alone does not satisfy AIVIA — affirmative consent must be obtained. You'll need a consent mechanism for AI video interviews used in Illinois, in addition to any LL144 notice.

Are there more AI hiring laws coming?

Yes. California has active legislation targeting algorithmic bias in hiring. Maryland's HB1202 is tracking AI hiring tool usage. Multiple states have AI governance bills under consideration that may include employment provisions. This regulatory environment will become more complex, not less. The advantage of building a compliance program now is that new state laws often borrow frameworks from existing ones — a solid LL144 compliance program gives you a foundation to build on.

Build a Multi-State Compliance Program With EmployArmor

EmployArmor provides purpose-built tools for each of the three primary LL144 requirements — the same requirements that form the core of multi-state AI hiring compliance:

• Bias Audit Tool — Independent auditor coordination, statistical analysis meeting NYC LL144 and Illinois AIVIA standards
• NYC LL144 Public Disclosure Page Tool — Compliant public disclosure generation and hosting
• Candidate Disclosure Notices Tool — Multi-jurisdiction notice generation covering NYC, Illinois, and Colorado requirements

For the full law text and compliance guide for each jurisdiction, see:

• NYC Local Law 144 law page
• Illinois AIVIA law page
• Colorado AI Act law page

Get your free multi-state compliance assessment →

Last updated: March 2026. This content is for informational purposes only and does not constitute legal advice. Consult qualified employment counsel familiar with each jurisdiction before implementing compliance programs. Statutory references: NYC Admin. Code § 20-870 et seq. (Local Law 144); 820 ILCS 42/1 et seq. (Illinois AIVIA); Colo. Rev. Stat. § 6-1-1701 et seq. (SB24-205). External resources: NYC Commission on Human Rights LL144 page; Illinois Attorney General; Colorado Attorney General.