EEOC Provides Guidance on Employers Using AI in the Workplace
By: Danielle Smid
The convenience and efficiency of utilizing Artificial Intelligence (“AI”) to hire employees, observe performance, and make decisions about pay and promotion opportunities could run your business into issues with Title VII, if the employer does not have sufficient oversight measures in place to monitor these tools effects. On May 18, 2023, the Equal Employment Opportunity Commission (“EEOC”) issued guidance on HR automated systems using AI and its impact on an employer’s Title VII compliance. The main concerns of this guidance focus on the selection procedures and whether they create a disparate impact that is prohibited by Title VII.
Algorithmic decision making has become more common throughout the employment process. These can include resume scanners that prioritize key words, software that monitors production, and job fit scores measuring cognitive and social skills as well as cultural fit. Employers utilize these AI tools in making employment decisions in all steps of the employment process. However, the EEOC is only focused on the “selection procedures” with the use of AI and the disproportionate effect the programs can have on the basis of race, color, religion, sex, or national origin, therefore being in violation of Title VII. The seemingly harmless AI assistance must be monitored to eliminate any disparate impact discrimination in the employment process. If a violation of Title VII were to be found in monitoring the AI systems, employers can show the use of the tool is job related and consistent with business necessity. Additionally, it will be important to determine if there is a less discriminatory alternative to implement in the selection process. Employers must also be aware that if the AI selection tools are administered by a third party, they may be held responsible for their agents such as in a situation where the agent has authority and is administering this program on behalf of the employer.
The new guidance does not create new policy but includes helpful tips employers can keep in mind when using AI selection tools. Employers can use the selection rate as a measurement to ensure compliance, which is the proportion of applicants or candidates who are hired or promoted and is calculated by dividing the number of people who are selected from the group by the total number of candidates in the group. The “four-fifths rule,” which was created in the 1978 Uniform Guidelines on Employee Selection Procedures, should be a general rule of thumb in determining whether the selection rate for one group is substantially different than the selection rate of another group, and it is substantially different if their ratio is less than four-fifths or 80%. Finally, if an employer discovers that the AI tool is creating a disparate impact in violation of Title VII, the employer can mitigate that impact by adjusting the tool to create an alternative algorithm or stop using it all together. If steps are not taken to mitigate the Title VII violation, that can open the employer to liability. Therefore, it is crucial that employers conduct self-audits on an ongoing basis to prevent a disproportionate selection effect prohibited under Title VII.
The new guidance indicates that the EEOC has paid more attention to discrimination issues with the use of advancing technology. It is important that employers using these types of AI tools remain diligent in their oversight to remain in compliance with Title VII in their selection processes.
If you have any questions, please contact Dani Smid or your BrownWinick attorney. Special thanks to summer associate Emily Bell for her assistance with this blog post.