Last month, New York City published final regulations (NYC 144) addressing the use of artificial intelligence by employers in hiring and promotion. The regulations prohibit employers and employment agencies from using automated employment decision tools (AEDTs) to make hiring and promotion decisions, unless the tool is audited for bias annually, the employer publishes a summary of the audit, and the employer provides notice to applicants and employees subject to the screening.
The regulations define an AEDT as any computational process, derived from machine learning, statistical modeling, data analytics, or artificial intelligence that issues a simplified output, including a score, classification, or recommendation, that is used substantially to assist or replace discretionary decision-making for employment decisions.
The bias audit must be conducted by an independent auditor that does not have financial or employment ties to the tool’s developer or the employer. Also, the annual audit must use data from the employer or employment agency’s historical use of the tool. If historical data does not exist or would not be statistically significant (e.g., because the sample size), the independent auditor may use historical data of other employers or employment agencies.
The bias audit aims to assess whether the AEDT has a disparate impact on a group of job applicants or employees based on a particular protected characteristic, such as sex, race, and ethnicity. The final rule states that the bias audit must calculate the “impact ratio” of each protected category. Although the employer is required to publish the impact ratios, it does not have to draw conclusions from the data.
The new regulations will take effect on July 5, 2023. By that date, New York City employers must determine whether they use an AEDT to make hiring or promotion decisions and, if so, it must (1) commission an independent bias audit; (2) publish a summary of the results of the audit; and (3) notify in advance applicants and employees of the tool’s use. Employers that violate the law, including its notice requirements, will be subject to fines of up to $500 for their first violation and up to $1,500 for each subsequent violation.
Several vendors currently offer AI-based hiring and promotion tools. Fletcher AI, for instance, represents that it can:
Automate the tedious parts of your recruiting process. Our advanced AI and human experience source high-quality candidates that match your requirements, eliminating the need for you to manually sift through resumes and job boards.
Fountain AI and HireVue offer similar AI-based recruiting and hiring tools. However, searches for vendors offering AEDT independent bias audits offer few options.
We anticipate that employers will take a “pause” on the use of AEDT hiring and promotion tools until they (1) can identify independent auditors with access to appropriate historical data; (2) determine whether the algorithms in AEDT have a disparate impact on protected classes; and (3) are prepared to publish impact ratios to applicants and employees. We recommend that employers retain and work through these steps with in-house or outside counsel so that their preliminary analyses and development of policies and procedures for compliance with this new law are protected in discovery.
Over the past two decades, federal government contractors regulated by the Office of Federal Contract Compliance Programs (OFCCP) have been subject to applicant tracking requirements, including the Internet Applicant Rule. Over the years, vendors have developed complaint applicant tracking systems that request and collect demographic information from candidates and hires that can be used by employers to generate hiring impact ratios like those required by the NYC regulations. We anticipate that these vendors (and the employers that rely on them) will be able to use those existing applicant tracking systems to audit AEDTs and provide the impact ratios required by the NYC regulations.
Stay tuned for additional developments in the new frontier of AI in the Workplace.