NYC Local Law 144 requires employers to conduct independent bias audits on automated employment decision tools. Learn what's required, who must comply, and how to ensure your organization meets all regulatory requirements.
The integration of AI into recruitment and hiring processes has revolutionized talent acquisition. From screening resumes to scheduling interviews and evaluating candidate fit, AI-powered tools promise unprecedented efficiency and objectivity. However, this technological advancement has also raised critical concerns about algorithmic bias and discrimination. New York City has taken a pioneering step to address these concerns through Local Law 144 (NYC LL 144), one of the most significant AI regulations affecting employers today.
This comprehensive guide explains everything you need to know about NYC LL 144, including what it requires, who it affects, and how to achieve compliance through proper bias auditing.
NYC Local Law 144, which took effect in 2023, regulates the use of "automated employment decision tools" (AEDTs) in recruitment and hiring processes. The law represents a landmark effort to ensure that AI systems used in employment decisions don't perpetuate or amplify discriminatory practices against protected groups.
AEDTs are any computational process, derived from machine learning, statistical modeling, data analytics, or artificial intelligence, that issues simplified output (such as scores, classifications, or recommendations) that are used to substantially assist or replace discretionary decision-making for employment decisions. This includes tools that:
Before diving into compliance requirements, it's worth understanding why organizations adopt these tools in the first place. AI-powered recruitment systems offer several compelling advantages:
Efficiency gains: Automating resume screening and initial candidate assessments can reduce time-to-hire by up to 70%
Enhanced objectivity: When properly designed, AI can reduce human biases by focusing on job-relevant qualifications
Improved candidate experience: Faster response times and streamlined processes create better experiences for applicants
Better talent matching: Advanced algorithms can identify qualified candidates who might be overlooked through manual screening
Cost reduction: Automation reduces the resource burden on HR teams, allowing them to focus on high-value activities
However, these benefits can only be realized when AI systems are fair, transparent, and free from discriminatory bias—precisely what NYC LL 144 aims to ensure.
NYC LL 144 establishes four fundamental requirements that employers must meet before using AEDTs:
Employers must ensure that an independent third party conducts a bias audit of any AEDT within one year before its use. This audit must be renewed annually to maintain compliance.
Who qualifies as an independent auditor? According to the NYC Department of Consumer and Worker Protection (DCWP), an independent auditor is a person or organization that:
Organizations must make information about the bias audit publicly available. This includes:
Employers must inform job candidates and employees about:
Candidates and employees must be provided with an alternative selection process or reasonable accommodation upon request.
The Equal Employment Opportunity Commission (EEOC) defines adverse impact as "a substantially different rate of selection in hiring, promotion, or other employment decision which works to the disadvantage of a race, sex, or ethnic group."
NYC LL 144 requires assessment of adverse impact across specific demographic categories to ensure AEDTs don't discriminate against protected groups.
The law mandates bias audits across the following categories:
Ethnicity categories:
Sex categories:
Intersectional categories:The law also requires testing of intersectional categories that combine sex and ethnicity (e.g., female Hispanic, male Asian, etc.), recognizing that discrimination can occur at the intersection of multiple protected characteristics.
A compliant bias audit must include:
Selection rates: The proportion of candidates from each demographic category who are selected (hired or promoted) by the AEDT
Impact ratios: A comparison of each group's selection rate to the most selected category, calculated as:
Impact Ratio = (Selection Rate for Group) / (Selection Rate for Most Selected Group)
Generally, an impact ratio below 0.80 (or 80%) indicates potential adverse impact, based on the EEOC's "four-fifths rule." However, statistical significance and practical considerations should also be evaluated.
Organizations that fail to comply with NYC LL 144 face significant consequences:
Given that each use of a non-compliant AEDT could constitute a separate violation, penalties can accumulate rapidly, creating substantial financial and reputational risks.
Begin by identifying all AEDTs currently in use or under consideration for recruitment and hiring processes. This includes:
Select a qualified independent auditor who:
Gather the necessary data for bias auditing, including:
The independent auditor will:
Make audit information available on your website or through other accessible means, including:
Establish processes to notify candidates and employees about:
If the audit reveals adverse impact:
Modern AI governance platforms can streamline the compliance process through automated bias testing capabilities. These solutions typically offer:
Automated data ingestion: Direct integration with HR systems and applicant tracking platforms to seamlessly collect necessary data
Statistical analysis: Automatic calculation of selection rates, impact ratios, and statistical significance testing across all required demographic categories
Compliance reporting: Generation of audit reports that meet regulatory requirements, including all mandated statistics and disclosures
Continuous monitoring: Ongoing tracking of AEDT performance to identify emerging bias issues before they become compliance violations
Documentation management: Centralized storage of audit reports, notifications, and compliance evidence
Beyond minimum compliance, leading organizations adopt these practices:
Don't wait until annual audits. Test for bias during:
Ensure AI systems are trained on representative datasets that reflect the diversity of your candidate pool. Unrepresentative training data is a primary source of algorithmic bias.
Maintain meaningful human review of AI recommendations, especially for final hiring decisions. AI should augment, not replace, human judgment in employment decisions.
Go beyond minimum notification requirements by:
When procuring third-party AEDTs:
Create a governance structure involving:
While NYC LL 144 is currently the most specific AI hiring regulation in the United States, it's part of a broader trend toward AI governance:
State-level initiatives: Several states are considering similar legislation for AI in employment
EU AI Act: Classifies certain AI systems in employment as "high-risk," requiring conformity assessments and transparency obligations
Federal proposals: Various federal bills addressing AI bias and discrimination are under consideration
Industry standards: Professional organizations are developing best practices and ethical guidelines for AI in HR
Organizations that establish robust AI governance practices now will be better positioned as this regulatory landscape evolves.
NYC Local Law 144 represents more than a compliance obligation—it's an opportunity to build trustworthy, fair, and effective AI-powered recruitment processes. Organizations that embrace rigorous bias testing and transparency don't just avoid penalties; they:
The pathway to compliance begins with understanding your obligations, conducting thorough bias audits through qualified independent auditors, implementing transparent notification procedures, and continuously monitoring your AEDTs for fairness. By partnering with experienced AI governance providers and adopting best practices that exceed minimum requirements, organizations can harness the power of AI in recruitment while ensuring equal opportunity for all candidates.
Navigating the complexities of NYC Local Law 144 requires more than just understanding the regulations—it demands robust technical capabilities, deep expertise in AI governance, and tools purpose-built for compliance at scale.
Holistic AI provides the complete solution for NYC LL 144 compliance and beyond.
Our comprehensive platform delivers everything you need to ensure your recruitment AI systems are compliant, fair, and trustworthy:
Automated Bias Testing & Auditing
AI System Discovery & Inventory
Continuous Monitoring & Risk Mitigation
Regulatory Compliance Automation
Red Teaming & Advanced Testing
Leading organizations including Unilever, MAPFRE, and Siemens trust Holistic AI to govern their AI systems responsibly. Our platform combines technical rigor with practical usability, enabling compliance teams, HR leaders, and data scientists to collaborate effectively.
Independent Expertise: As an independent third-party auditor, we meet NYC LL 144's requirements while bringing objectivity to your compliance efforts
Research-Driven Approach: Our solutions are grounded in cutting-edge AI safety research and academic partnerships with institutions like UCL
Proven Track Record: Demonstrated technical credibility through competitive achievements and real-world deployments with major enterprises
Comprehensive Coverage: Unlike point solutions, we address the entire AI lifecycle—from discovery and inventory to testing, monitoring, and ongoing governance
Future-Proof Platform: Built to scale with your AI adoption and adapt to evolving regulations like the EU AI Act and emerging US federal requirements
Don't let compliance concerns slow down your AI-powered recruitment initiatives. With Holistic AI, you can confidently deploy AEDTs knowing that:
Transform compliance from a checkbox exercise into a competitive advantage.
Contact Holistic AI today to schedule a demo and discover how our AI governance platform can help you achieve NYC LL 144 compliance while building more effective, trustworthy recruitment systems.
Ready to ensure your recruitment AI is compliant and fair?
Visit holistic.ai or reach out to our team to learn how we can support your responsible AI journey.