What Is the NYC AI Hiring Law? A Detailed Guide to New York City Local Law 144 Compliance

Currently In Effect

Regulates AI-Driven Hiring Tools

Mandates Annual 3rd Party Bias Audits

Requires Disclosure Of AEDT Use

AI for Regulatory Compliance

FAQs About New York City’s Local Law 144

Steps to Achieve Compliance with Local Law 144

NYC Local Law

High-Level Summary

  • NYC Local Law 144 prohibits employers from using automated employment decision tools (AEDTs) without an independent auditor assessing the AEDT for bias.
  • Bias audits must be completed and published every 12 months.
  • Employers must notify candidates that they use an AEDT, which qualifications the AEDT assesses, the types and sources of data the business collects for the AEDT, and its data retention policy.
  • Employers must also provide the candidates with an opportunity to request an alternative selection process or accommodation, if available.

Who does Local Law 144 apply to?

NYC Law 144 applies to all employers and employment agencies that hire employees who reside in New York City, regardless of their location.

Companies based in NYC or evaluating applicants in “the City” must comply with the law.

That means that any company that employs or considers employing individuals residing in the city must comply with Local Law 144.

In addition to the broad geographic scope, the law targets the use of Automated Employment Decision Tools (AEDTs).

Currently, an AEDT is defined as a tool that provides a simplified output (say, a model score) to be used as the only or most significant criterion in a selection process or as an override to a human decision.

Compliance Requirements of NYC LL 144

The requirements are composed of three main parts:

1. Independent audit: Employers are required to engage an independent third party to conduct a bias audit of their tool. These audits must be completed annually.

This audit is based on the AEDT’s differential impact on different demographic groups, specifically on gender and race/ethnicity, and for all the possible intersections of the two – e.g. “Hispanic women”, “Asian men”. For AEDTs that make a binary pass/fail decision, the audit must assess the selection rate of each category, defined as the number of candidates passed divided by the number of candidates total within that group.

For AEDTs with continuous score outputs, the audit must use a scoring rate, which is calculated similarly to the selection rate and describes how many candidates received a score greater than the median score (the median across the entire set of candidates).

The audit must then calculate the impact ratio, or the ratio between each group’s selection (or scoring) rate and the group with the highest selection (or scoring) rate. In the spirit of the four-fifths rule, impact ratios of less than 80% could potentially signal bias against the affected group. While the bill does not explicitly prohibit employers from using an AEDT that breaks the four-fifths rule, discrimination in hiring is illegal per federal, state, and NYC law.

2. Notice to candidates: Employers must also give candidates at least 10 business days’ notice before using AEDTs, with an option to request an alternative assessment method, if reasonable.

3. Results visibility: Employers must also publish the date and the results of the most recent AEDT audits.

*Regarding small sample sizes for the assessed categories, the law permits auditors to exclude categories composing less than 2% of the data from the bias analysis.

Non-Compliance Penalties

Penalties for non-compliance are a $500 fine for the first violation and $1,500 fines for subsequent violations.

Status

The bill has been passed and enforcement started on July 5, 2023.

How does New York LL144 compare to NY A00567?

How can companies best implement LL 144’s requirements?

There are three major considerations for companies regarding this regulation:

1. Engage a Third-Party Auditor: Partner with a qualified external vendor who specializes in evaluating the fairness and efficacy of AEDTs (such as FairNow). This third party should conduct annual bias audits (at a minimum), ensuring the tools meet the regulatory requirements for impartiality and non-discrimination.

2. Prioritize Transparency & Communication: Maintain transparency by notifying applicants about the use of AEDTs in their evaluation process. Additionally, publicize the results of the third-party audits and the methodologies used on your company’s website, as mandated by LL 144.

3. Document Compliance and Continuous Improvement: Keep detailed records of the evaluation processes, findings, and any actions taken to address disparities. These insights can be used to improve the AEDT systems continuously, ensuring they remain compliant with LL 144 and other applicable regulations over time.

By focusing on these three steps—engaging a third-party vendor, ensuring transparency, and documenting compliance and improvements—companies can effectively meet the requirements of LL 144 while ensuring fairness in their employment practices.

Organizations looking to stay ahead of the curve and gain a competitive edge should keep a close eye on developments related to NYC’s LL144 readiness this year.

Staying informed and engaged will be key to achieving AI compliance excellence.

Not sure you have the time to sift through hundreds of pages of new legislation? Luckily, we already do that for fun! Add your email to the form below and we’ll keep you informed anytime new, meaningful legislation breaks.

AI Model Governance

AI Compliance doesn’t have to be so complicated.

Use FairNow’s AI governance platform to:

Effortlessly ensure your AI is in harmony with both current and upcoming regulations

Ensure that your AI is fair and reliable using our proprietary testing suite

Stay ahead of compliance requirements before fees and fines become commonplace

Get Expert Help With AI Governance

Schedule a free consultation with our AI governance experts today.