NYC Local Law 144 (2021)
Complete Legal Breakdown with Rules, Layman Explanations & Citations
Who This Law Applies To
Geographic Scope
Citation: NYC Admin. Code § 20-870 et seq.
Applies to:
- Employers with offices in NYC
- Employment agencies operating in NYC
💡 Layman Explanation:
If you're hiring for a job in NYC, or considering NYC employees for promotion, this law applies to you.
What's Covered
✅ Covered:
- • Hiring decisions
- • Promotion decisions
❌ Not Covered:
- • General HR processes
- • Performance reviews (unless tied to promotion)
- • Outreach to potential candidates
Citation: NYC Admin. Code § 20-870 "Employment decision means to screen candidates for employment or employees for promotion within the city."
Key Definitions in Plain English
1. Automated Employment Decision Tool (AEDT)
Legal Definition:
"Any computational process, derived from machine learning, statistical modeling, data analytics, or artificial intelligence, that issues simplified output, including a score, classification, or recommendation, that is used to substantially assist or replace discretionary decision making for making employment decisions that impact natural persons."
Citation: NYC Admin. Code § 20-870
💡 Layman Explanation:
Software that uses AI/algorithms to help make hiring or promotion decisions by giving scores, rankings, or recommendations.
What COUNTS as an AEDT:
- ✅Resume screening software that scores candidates 1-100
- ✅Video interview AI that analyzes facial expressions/speech
- ✅Personality test software that recommends "good fit" candidates
- ✅Skills assessment tools that rank applicants
- ✅Chatbot screeners that filter candidates before human review
❌What DOES NOT count:
- ❌Junk email filters (doesn't impact people directly)
- ❌Spell-checkers (not making employment decisions)
- ❌Spreadsheets/databases (just storing data)
- ❌ATS systems that only organize applications without scoring
- ❌Calendaring tools for interview scheduling
🔑 Key Test:
Does it give a simplified output (score, yes/no, ranking) that substantially helps make the hiring decision?
2. Bias Audit
Legal Definition:
"An impartial evaluation by an independent auditor... testing of an automated employment decision tool to assess the tool's disparate impact on persons of any component 1 category required to be reported by employers pursuant to subsection (c) of section 2000e-8 of title 42 of the United States code."
Citation: NYC Admin. Code § 20-870
💡 Layman Explanation:
An independent expert tests your AI tool to see if it unfairly disadvantages certain groups of people.
What Gets Tested:
Must test for disparate impact on:
Sex:
- • Male vs. Female
Race/Ethnicity (EEOC EEO-1):
- • Hispanic or Latino
- • White (Not Hispanic or Latino)
- • Black or African American
- • Asian (Not Hispanic or Latino)
- • American Indian or Alaska Native
- • Native Hawaiian or Pacific Islander
- • Two or More Races
Plus all intersections (e.g., Hispanic Female, White Male, etc.)
Citation: RCNY § 5-300 - References EEOC categories
Three Core Requirements
Citation: NYC Admin. Code § 20-871
Annual Bias Audit
Tool must be audited within the last 12 months before use
Public Disclosure
Audit results must be posted publicly on your website
Candidate Notice
Notify applicants/employees about AEDT use (at least 10 business days before)
💡 Layman Summary:
You must (1) test for bias yearly, (2) publish the results, (3) tell people you're using AI.
Penalties & Enforcement
Enforcement Authority
Citation: NYC Admin. Code § 20-873
Enforcer: NYC Department of Consumer and Worker Protection (DCWP)
Penalty Structure
Citation: NYC Admin. Code § 20-872 & RCNY § 6-81
First Violation
$500
per violation
Subsequent Violations
$500-$1,500
per violation
⚠️ "Per Violation" Defined:
- Type 1: Each day using AEDT without required bias audit
- Type 2: Each failure to notify a candidate
- Type 3: Each failure to publish audit results
Daily Accumulation: If you use an AEDT without a valid audit for 30 days, that's 30 violations.
Compliance Checklist
PRE-DEPLOYMENT (Before Using Any AEDT)
- Identify if tool is an AEDT - Does it score, rank, or recommend candidates?
- Conduct bias audit - Hire independent auditor, use historical data, test all categories
- Publish audit results - Post on website, make publicly accessible, include distribution date
- Prepare candidate notices - Draft templates, include in job postings, set up automation
DURING USE
- Provide notice to candidates - At least 10 business days before AEDT use
- Respond to data requests - Within 30 days, provide data type, source, retention policy
- Monitor tool use - Track notifications, maintain records for audit compliance
ANNUAL MAINTENANCE
- Conduct new bias audit - Within 12 months of last audit
- Update published results - Post new audit results on website
- Review and update notices - If tool changes, update candidate notifications
Official Sources & Citations
Informational Resource
This resource provides general information about NYC Local Law 144 and is designed to help you understand the requirements. The information is compiled from publicly available sources and official legislative text. For specific legal advice tailored to your situation, we recommend consulting with a qualified attorney or compliance professional.