This month, the U.S. Equal Employment Opportunity Commission (“EEOC”) issued its technical assistance document  to provide guidance to employers on the  American with Disabilities Act  (ADA), and the use of software, algorithms and artificial intelligence (collectively “AI tools”) to assess job applicants and employees.   The EEOC notes that employers have a responsibility to inspect AI tools for disability bias and should have plans to provide reasonable accommodations. 

Employers may rely on different types of AI tools at several stages of the employment process. Examples include: resume scanners that prioritize applications using certain keywords; employee monitoring software that rates employees on the basis of their keystrokes or other factors; “virtual assistants” or “chatbots” that ask job candidates about their qualifications and reject those who do not meet pre-defined requirements; video interviewing software that evaluates candidates based on their facial expressions and speech patterns; and testing software that provides “job fit” scores for applicants or employees regarding their personalities, aptitudes, cognitive skills, or perceived “cultural fit” based on their performance on a game or on a more traditional test.

The EEOC guidance noted promising practices that employers may take to comply with the ADA when using AI tools that include:

• Employers must provide reasonable accommodations when legally required (e.g., training staff to recognize and process requests for reasonable accommodation as quickly as possible, including requests to retake a test in an alternative format, or be assessed in a different way, after the individual has already received poor results).
• Employers should minimize the chances that the AI tools would disadvantage individuals with disabilities, either intentionally or unintentionally (e.g., using AI tools that have been designed to be accessible to individuals with as many different kinds of disabilities as possible; informing applicants and employees that are being rated that reasonable accommodations are available for individuals with disabilities,  and provide clear and accessible instructions for requesting such accommodations).
• Employers may seek to minimalize the changes that AI tools would assign poor ratings to individuals who are able to perform the essential functions of the job, with reasonable accommodations if one is legally required (e.g., ensure the AI tools only measure abilities or qualifications that are truly necessary for the job—even for people who are entitled to an on-the-job reasonable accommodation).
• Before purchasing an AI tool, an employer should ask the vendor to confirm that the tool does not ask the job applicants or employees questions that are likely to elicit information about a disability or seek information about an individual’s physical or mental impairments or health, unless such inquiries are related to a request for reasonable accommodation.

In addition to the ADA, AI tools have been found to also inadvertently cause disparate treatment based on other protected classes (see Amazon Scraps Secret AI Recruiting Tool That Showed Bias Against Women). For example, setting up an AI tool to screen for a college degree as a job requirement may adversely impact groups with a lower college-completion rate, though the individual applicants may be otherwise qualified for the job.

AI tools can be tremendous time-savers and eliminate much of the implicit bias that may come from human screening processes. However, whether an employer screens their applicants using an AI tool or not, they should always be mindful of where applicants may need accommodations and where other groups may fall (or be kicked) out of the process.