Foster Garvey -- EEOC Releases New Guidance for AI Algorithms in Employment Decisions

  • /sites/default/files/styles/cover/public/cover/random/2017-11/cover-pic-02.jpg?h=afa3cfa7&itok=QvEihQ2y
  • /sites/default/files/styles/cover/public/cover/random/2017-11/cover-pic-03.jpg?h=452f395a&itok=o2eJpQ1X
  • /sites/default/files/styles/cover/public/cover/random/2017-11/cover-pic-04.jpg?h=d85646e8&itok=e-zcRWuw
  • /sites/default/files/styles/cover/public/cover/random/2017-11/cover-pic-05.jpg?h=eb90c5f1&itok=fmftIU1H
  • /sites/default/files/styles/cover/public/cover/random/2017-11/cover-pic-06.jpg?h=f8567693&itok=OYoPjORc

from Mark Smith, Smith Policy Group

On May 18, 2023, the EEOC released guidance on the use of Artificial Intelligence (“AI") tools in employment decisions. Though primarily focused on the selection and hiring of candidates, the same general guidance on AI tools would apply to their use in all employment decisions or personnel actions. The bulletin is short and worth reading.

The EEOC’s guidance on the topic should hopefully not be surprising to employers, as it follows established disparate impact analysis. More specifically, when using AI tools, employers should be cautious of using selection procedures that have a disproportionately negative effect on employees or applicants who have characteristics protected by Title VII.

The EEOC guidance also highlights some key reminders for employers that use algorithmic decision-making tools for selection and hiring:

  • Employers should conduct ongoing analysis of any of their AI tools to determine whether their employment practices have a disparate impact.
  • Employers may still be liable even if a problematic test was developed or implemented by a third party on their behalf (such as a recruiting agency).
  • Employers should make direct inquiries to any third party as to the potential disparate impacts of any AI tools used.
  • If in deciding on which tools to use, failure to adopt a less discriminatory algorithm considered during the development process may give rise to liability.

The takeaway from this guidance is that employers should not turn a blind eye to the use of AI tools by their own employees or any third party or agent acting on their behalf. Direct inquiry, analysis and confirmation that any AI tool usage for employment decisions does not result in disparate impact is critical and should be periodically reviewed.



Smith Policy Group
1001 K Street, 6th Floor
Sacramento, CA 95814
(916) 335-5072