New York City skyline
NYC Local Law 144

NYC automated employment decision tools compliance

NYC Local Law 144 requires bias audits and notice for automated hiring tools used in New York City. Effective July 5, 2023, with penalties up to $1,500 per violation per day.

Book compliance consultation

What is NYC Local Law 144?

NYC Local Law 144 of 2021 (Int 1894-2020) regulates the use of Automated Employment Decision Tools (AEDTs) by employers and employment agencies in New York City. It requires annual bias audits, candidate notification, and public disclosure of audit results.

Effective July 5, 2023: Employers using AI, machine learning, or algorithmic tools to screen, evaluate, or select candidates for NYC positions must comply with bias audit and notice requirements. The law is enforced by the NYC Department of Consumer and Worker Protection (DCWP).

Bias audit

Annual independent audit required

Notice

10-day advance candidate notification

Complements EU AI Act employment provisions and FTC AI guidelines.

Who needs to comply?

Employers in NYC

Organizations using AEDTs to screen or hire candidates for NYC positions

Employment agencies

Agencies using AEDTs to evaluate or refer NYC-based candidates

Software vendors

AEDT providers selling to employers or agencies in NYC

Remote hiring teams

Organizations hiring for remote roles but located in NYC

Promotion decisions

Employers using AEDTs for internal promotion or transfer in NYC

Multi-state operations

National employers with NYC employees or hiring for NYC positions

Want to see how your audit report looks?

Download a sample NYC Local Law 144 bias audit report generated by VerifyWise.

Download sample (PDF)

How VerifyWise supports Local Law 144 compliance

VerifyWise covers the core quantitative requirements of LL144: bias audit computation, demographic category analysis, intersectional breakdowns, and compliance record-keeping

LL144 requirement
VerifyWise coverage
Annual bias audits
Dedicated bias audit module with full audit lifecycle: create, run, review results, and archive
Selection rate analysis
Engine computes per-group selection rates across all demographic groups
Impact ratio (four-fifths rule)
Automatic impact ratio calculation with configurable 0.80 threshold and flag detection
EEO Component 1 categories
Pre-configured preset with sex (2 groups) and race/ethnicity (7 groups)
Intersectional analysis
Automatic sex-by-race cross-tabulation covering all 14 combinations
Small sample exclusion
Groups under 2% of total applicants excluded from ratio calculations per DCWP guidance
Audit metadata
Captures AEDT name, distribution date, and data source description
Compliance documentation
Full audit trail with timestamped results, configuration snapshots, and per-group breakdowns
Data transparency
AI Trust Center for public disclosure of audit summaries, satisfying the publication requirement
Results export
JSON download of complete audit results for sharing with auditors or regulators

Beyond the bias audit

AEDT identification and inventory

Systematically identify all automated employment decision tools in your organization. The platform tracks each tool's purpose, vendor, deployment status and jurisdiction coverage to ensure you know exactly which systems fall under Local Law 144.

Addresses: Scope determination: What constitutes an AEDT

Bias audit tracking and coordination

Manage the annual bias audit requirement with workflow tracking for independent auditor selection, data provision and audit completion. The platform maintains audit history and flags upcoming deadlines to ensure continuous compliance.

Addresses: Bias audit: Annual independent auditor requirement

Disparate impact documentation

Document disparate impact testing results across required categories (race/ethnicity and sex). The platform stores selection rates, impact ratios and statistical analyses that satisfy DCWP reporting requirements.

Addresses: Bias audit: Disparate impact testing and metrics

Notice and transparency management

Track candidate notification requirements with automated reminders 10 business days before AEDT use. The platform documents notice delivery, method and timing to demonstrate compliance with transparency obligations.

Addresses: Notice: 10-day advance notification to candidates

Publication and disclosure tracking

Maintain public website publication requirements for bias audit results, methodologies and distribution dates. The platform tracks when results were published, updated and ensures all required information remains publicly accessible.

Addresses: Publication: Audit results on employer website

Alternative process and data retention

Document alternative selection processes and maintain required data retention. The platform stores historical audit data, exemption justifications and demonstrates you've provided alternatives as the law requires.

Addresses: Alternatives: Alternative selection process availability

All compliance activities are tracked with timestamps, audit trails, and evidence documentation. This demonstrates systematic compliance rather than after-the-fact documentation.

Complete Local Law 144 requirements coverage

VerifyWise provides dedicated tooling for all core compliance requirements

11

Compliance requirements

11

Requirements with dedicated tooling

100%

Coverage across all obligations

Bias audit4/4

Annual independent audits with disparate impact testing

Notice requirements3/3

Candidate notification 10 days before AEDT use

Publication2/2

Audit results and methodology on public website

Data retention2/2

Historical data and alternative selection process records

Built for NYC employment AI compliance

Annual audit tracking

Automated reminders and deadline management for renewals

Disparate impact tools

Calculate selection rates and impact ratios by category

Notice automation

Track 10-day notice delivery with documentation

Multi-jurisdiction support

Crosswalk to EU AI Act and other employment AI laws

Bias audit requirements

What must be tested and how to comply with independent audit obligations

Independent auditor

Bias audit must be conducted by an independent auditor who did not develop or is not currently marketing the AEDT.

  • Cannot be tool developer or vendor
  • Must be impartial third party
  • Documented independence criteria

Disparate impact testing

Calculate selection rates and impact ratios for each category (race/ethnicity and sex) to determine disparate impact.

  • Selection rate by category
  • Impact ratio calculations
  • Statistical significance testing

Required categories

Testing must cover race/ethnicity categories and sex categories as defined by DCWP regulations.

  • Hispanic or Latino
  • White (Not Hispanic or Latino)
  • Black or African American (Not Hispanic or Latino)
  • Asian (Not Hispanic or Latino)
  • Female
  • Male

Annual audit frequency

Bias audit must be completed within one year prior to the use of the AEDT and updated annually.

  • Audit valid for 1 year
  • Must renew before expiration
  • Track audit completion dates

Disparate impact testing methodology

The bias audit must calculate the selection rate for each category (e.g., Hispanic or Latino, Female) and compare it to the selection rate of the most selected category. An impact ratio below 0.8 (80% rule) may indicate disparate impact under EEOC guidelines.

Selection rate

(Selected from category) / (Total from category)

Impact ratio

(Category selection rate) / (Highest selection rate)

Notice and transparency requirements

What candidates and employees must be told and when

Candidate notification

At least 10 business days before use of AEDT

  • Notice that AEDT will be used
  • Job qualifications and characteristics used by AEDT
  • Data source and retention policy
  • Method of notice (written or electronic)

Employee notification

For AEDTs used on current employees

  • Notice before use for promotion/transfer
  • Same 10-day advance notice period
  • Job qualifications used by tool
  • Employee data retention practices

Publication requirements

Public website disclosure

  • Bias audit summary results
  • Distribution date of AEDT
  • Auditor name and methodology
  • Instructions for requesting alternative process

Alternative selection process requirement

Employers must provide a process for candidates and employees to request an alternative selection process or reasonable accommodation as an alternative to the AEDT. Instructions for making this request must be published on the employer's website alongside bias audit results.

What qualifies as an AEDT?

Common automated employment decision tools covered by Local Law 144

Resume screening

  • AI resume parsers
  • Keyword matching systems
  • Applicant ranking tools

Interview assessment

  • Video interview analysis
  • Speech pattern evaluation
  • Facial expression analysis

Testing & evaluation

  • Automated skills tests
  • Personality assessments
  • Game-based assessments

Internal promotion

  • Performance evaluation AI
  • Succession planning tools
  • Internal mobility platforms

Definition: Any computational process derived from machine learning, statistical modeling, data analytics, or AI that substantially assists or replaces discretionary decision-making for hiring or promotion.

If a tool scores, ranks, classifies, or recommends candidates, it likely qualifies as an AEDT.

12-week compliance roadmap

A practical path to Local Law 144 compliance with clear milestones

Phase 1Weeks 1-3

Discovery & inventory

  • Identify all AEDTs in use or planned
  • Determine NYC jurisdiction applicability
  • Document current vendor relationships
  • Assess current notice and audit status
Phase 2Weeks 4-8

Bias audit planning

  • Select independent auditor
  • Establish data collection processes
  • Define historical data requirements
  • Create audit timeline and milestones
Phase 3Weeks 9-12

Notice & transparency

  • Develop candidate notification process
  • Create public website disclosure page
  • Implement 10-day notice tracking
  • Document alternative process options
Phase 4Continuous

Ongoing compliance

  • Schedule annual bias audits
  • Monitor notice delivery compliance
  • Update public disclosures
  • Maintain audit and notice records
Enforcement & penalties

Penalties for non-compliance

The NYC Department of Consumer and Worker Protection (DCWP) enforces Local Law 144 with significant daily penalties

First violation

$500 per violation

Initial non-compliance with notice, audit, or publication requirements

Subsequent violations

$500-$1,500 per violation

Repeated failures to comply with any Local Law 144 requirement

Per-day violations

Each day = separate violation

Ongoing non-compliance accrues daily penalties until resolved

Example penalty calculation: Failing to conduct a bias audit for 30 days could result in penalties of $15,000 to $45,000 (30 days × $500-$1,500 per day). Continuous compliance is essential.

Enforced by: NYC Department of Consumer and Worker Protection

Policy templates

AI governance policy repository

Access ready-to-use employment AI policy templates aligned with Local Law 144, EU AI Act and other regulations

Bias audit policies

  • • Bias Audit Policy
  • • Independent Auditor Selection
  • • Disparate Impact Testing
  • • Audit Results Publication
  • • Annual Audit Schedule
  • + 3 more policies

Notice & transparency

  • • Candidate Notice Policy
  • • Employee Notification
  • • Website Publication Requirements
  • • Alternative Process Policy
  • • Data Retention Standards
  • + 2 more policies

AEDT governance

  • • AEDT Inventory Management
  • • Vendor Compliance Requirements
  • • Employment AI Governance
  • • Fairness & Non-Discrimination
  • • Compliance Monitoring
  • + 4 more policies
Browse all policy templates

Frequently asked questions

Common questions about NYC Local Law 144 compliance

An AEDT is any computational process, derived from machine learning, statistical modeling, data analytics, or artificial intelligence, that issues simplified output, including a score, classification, or recommendation, to substantially assist or replace discretionary decision-making for employment decisions that impact candidates or employees. See the NYC DCWP official page for complete definitions.
NYC Local Law 144 became effective on July 5, 2023. Employers and employment agencies using AEDTs in NYC must comply with all notice, bias audit, and publication requirements as of this date. Penalties apply for violations occurring after the effective date.
A bias audit must be conducted by an independent auditor within one year before AEDT use. The audit must test for disparate impact across race/ethnicity categories (Hispanic or Latino, White, Black or African American, Asian) and sex categories (Male, Female). Selection rates and impact ratios must be calculated and published on your website.
Employers must provide candidates with notice at least 10 business days before using an AEDT. The notice must state that an AEDT will be used, the job qualifications and characteristics the AEDT will assess, and information about data collection and retention. Notice can be provided in writing or electronically.
Employers must publicly post on their website: (1) the date the AEDT was used, (2) the bias audit summary results including selection rates and impact ratios by category, (3) the distribution date of the AEDT version audited, and (4) instructions for requesting an alternative selection process or reasonable accommodation.
Bias audits must be conducted at least once per year. A bias audit is valid for one year from the date of completion. If you continue using the AEDT beyond one year, you must complete a new bias audit before the previous one expires.
The bias audit must be conducted by an independent auditor. The auditor cannot be the developer of the AEDT, cannot be currently marketing the AEDT, and must be free from conflicts of interest. Many organizations hire specialized AI audit firms or consulting firms with bias testing expertise.
Penalties range from $500 to $1,500 per violation. Each day a violation continues is considered a separate violation. The NYC DCWP enforces compliance and can assess penalties for failures to conduct bias audits, provide required notices, or publish audit results.
Yes, if the position can be performed in NYC or the employer is located in NYC. The law applies based on where the employment is or can be performed. If you're hiring for a remote role that can be performed from NYC, or if your company is based in NYC, Local Law 144 likely applies.
The employer or employment agency using the AEDT remains responsible for compliance even when using vendor tools. You must ensure the vendor can provide or support bias audits, that you can provide required candidate notices, and that you can publish the required information on your website. Include compliance requirements in vendor contracts.
Very limited exemptions exist. The law exempts AEDTs that don't substantially assist or replace discretionary decision-making. If a tool only exports data without ranking, scoring, or recommending candidates, it may not meet the AEDT definition. However, most AI-powered hiring tools do fall under the law's scope.
NYC Local Law 144 is a specific municipal employment law. It complements broader frameworks like the EU AI Act (which also regulates employment AI) and FTC AI guidelines on algorithmic fairness. Organizations may need to comply with multiple overlapping requirements.
Employers must retain bias audit records, notices provided to candidates, published website disclosures, and data about alternative selection processes offered. Historical data used for bias audits (selection rates by category) should be maintained to demonstrate ongoing compliance and audit trail.
Candidates and employees must be able to request an alternative selection process or reasonable accommodation as an alternative to the AEDT. This could include human-only review, different assessment methods, or accommodations for disabilities. You must provide instructions for making such requests on your website.
Yes, VerifyWise helps you inventory AEDTs, track bias audit requirements and deadlines, document notice delivery, maintain publication records, and generate compliance reports. Our platform also provides crosswalks to related regulations like the EU AI Act for organizations operating across multiple jurisdictions.
The four-fifths rule comes from EEOC adverse impact analysis. If any demographic group's selection rate is less than 80% of the highest-selected group's rate, that's a flag for potential adverse impact. LL144 requires auditors to calculate and report these impact ratios by race, ethnicity and sex. Falling below 0.80 doesn't automatically mean a violation, but it's the threshold that gets regulators looking more closely.
Selection rate is the most common metric: the probability that an applicant gets advanced, recommended or selected by the system. Other useful metrics include false positive rate (advancing unqualified candidates), false negative rate (filtering out qualified ones), true positive rate and score distributions across demographic groups. LL144 specifically requires impact ratios based on selection rate or scoring rate (the rate at which a group scores above the overall median).
The best case is real applicant data with self-declared demographic information. Some vendors get this through employer ATS data-sharing arrangements or their own consent-based intake process. When demographic data isn't available (which is common), LL144 allows test data, including inferred demographic characteristics or synthetic representative datasets. Either approach requires documented methodology and compliance with applicable privacy law.
ISAE 3000 is the International Standard on Assurance Engagements for non-financial subject matter. It's the same framework accounting firms use for non-financial assurance work. When applied to AI audits, it means the engagement follows defined standards for evidence gathering, reporting and auditor independence. The result is a report that holds up to regulatory scrutiny and enterprise due diligence, unlike audits conducted without a formal standard.
Regulators, procurement teams and courts treat self-reported testing very differently from third-party verified results. An independent auditor has no financial or organizational relationship to the AI system being evaluated. Under ISAE 3000, independence is a formal requirement. An audit conducted by the vendor's own team, or by a firm with a consulting relationship to the vendor, won't hold up well in a regulatory inquiry.
DCWP can issue fines for non-compliance under LL144. After the December 2025 Comptroller audit, DCWP has taken a more proactive enforcement stance, using EEOC and FTC filings as leads for enforcement inquiries. Large employers with high-volume NYC hiring face the most risk. Beyond fines, missing audit documentation is becoming a deal-breaker in enterprise procurement; vendor questionnaires now routinely ask for it.
The EU AI Act classifies AI systems used in employment and worker management as high-risk. These systems must go through conformity assessments covering technical documentation, risk management, data governance and human oversight. The EU AI Act doesn't use the term "bias audit," but the conformity assessment requirements overlap heavily with what a rigorous independent audit covers. Learn more about EU AI Act compliance.

Related reading

NYC Local Law 144 compliance: A practical checklist for employers using AI in hiring

A step-by-step guide covering AEDT inventory, bias audits, candidate notification, and what the December 2025 Comptroller's audit means for enforcement.

Ready to achieve Local Law 144 compliance?

Start managing your bias audits, candidate notices, and ongoing compliance requirements today.

Schedule compliance consultation
NYC Local Law 144: AEDT bias audit and notice requirements