On January 1, 2020, Illinois' new Artificial Intelligence Video Interview Act (AIVIA) went into effect, meaning Illinois employers must now comply with the law if they use artificial intelligence (AI) to analyze video interviews by job candidates. As we outlined in a prior post, the AIVIA imposes duties of transparency, consent and data destruction on organizations using AI to evaluate interviewees for jobs that are "based" in Illinois.
While these concepts may be clear in the abstract, the Illinois law is a lesson in brevity and leaves several key terms undefined (including, for example, the term "artificial intelligence"). Nor is it clear what it means for a position to be "based" in Illinois. As a result, employers using AI-enabled analytics in interview videos must sort through these questions and take other affirmative steps to ensure compliance with the new law.
Recommended Steps for Complying with the AIVIA
As an initial step, employers using AI interview technology for jobs based in Illinois should undertake a scoping analysis to examine their current reliance on AI-enabled technology in order to determine their potential obligations under the new law. If using a third-party service provider that offers pre-employment video assessment tools, employers must determine if that vendor’s product utilizes AI. If so, employers should also review the terms of service with that vendor to identify (or, if necessary, develop) terms that allocate responsibility and potential liability for complying with the mandates of the AIVIA.
In addition, employers using AI-enabled tools for video interviews, whether through a third-party vendor or on their own, must take several affirmative steps to ensure compliance with the AIVIA’s requirements:
Notify the applicant in advance that the organization is using AI-enabled technology to analyze video interviews.
- The best practice is to inform applicants in writing, and to retain a record of the notice.
Explain to the applicant "how the [AI] works" and what "general types of characteristics the technology uses to evaluate applicants." To the extent the employer is using a vendor’s software, it can use the vendor's public-facing description of how its product works. However, it is important that employers have a firm understanding of the vendor's technology and be sure it is disclosed properly to the applicant. Also significant is the law’s requirement that entities explain the "general types of characteristics" the AI uses to evaluate applications.
- If, for example, the technology analyzes facial expressions or other biometric information, that should be disclosed (which may also trigger additional obligations under Illinois' Biometric Information Privacy Act).
Obtain in advance the applicant's consent to use the AI "as described in the information provided." This limitation means that employers must ensure that their disclosures accurately describe how the AI works and the types of characteristics it relies upon to evaluate applicants.
- If the disclosure is not accurate or is insufficient, that could give rise to later claims that the disclosure was insufficient and/or that the consent was not fully informed. Again, the best practice is to have a written record of the applicant’s consent.
4. Limits on Distribution
Employers must limit the distribution and sharing of videos to only those persons "whose expertise or technology" is necessary to evaluate an applicant's fitness for the position.
- It is a good practice to establish written policies with respect to limiting the sharing of these videos in the same way that entities utilize such policies for limiting the sharing of other sensitive private files or data of employees.
5. Duty to Destroy
Upon request from the applicant, destroy the video (and all backup copies held by the employer, its vendors or any other party) within 30 days.
- It remains unclear whether there is a conflict between the AIVIA's duty to destroy and other state and federal record preservation laws, such as the U.S. Equal Employment Opportunity Commission (EEOC) regulation requiring employers to retain "any personnel or employment record," including "application forms submitted by applicants and other records having to do with hiring," for a year or more.
Whether the applicant interview videos constitute "records" under the EEOC regulation, or whether the AIVIA requires employers to destroy any data or analytics generated by the AI beyond just destroying the video, is not addressed in the AIVIA and remains an open question.
If you have questions concerning compliance with the AIVIA or the use of automated decision-making in your company's personnel practices, members of Davis Wright Tremaine's Employment Services Group and AI Team can provide the necessary guidance to mitigate any risks that may arise.
Recent FTC Complaint Demonstrates Increased Scrutiny of AI Video Interview Services
This niche technology has garnered even more scrutiny in the last few months. On November 6, 2019, the Electronic Privacy Information Center (EPIC) filed a complaint with the Federal Trade Commission (FTC) against video interview technology company HireVue. EPIC alleged that HireVue’s analytics platform was unfair and deceptive.
HireVue conducts video-based and game-based "pre-employment" assessments of job candidates on behalf of potential employers. The process works by collecting "tens of thousands of data points" from each video interview of a job candidate, including a candidate’s intonation, inflection, and emotions, and then using predictive algorithms to determine each job candidate’s employability.
EPIC alleges HireVue falsely represents to job candidates that it does not use facial recognition technology or track facial features for identity recognition purposes because it collects candidate facial data to evaluate expressions of emotion and personality. EPIC claims this violates the Federal Trade Commission Act (FTC Act). EPIC also alleges HireVue's use of biometric data and algorithms is "unfair" under the FTC Act because its "secret, 'predictive algorithms'" make it impossible for job candidates to know how their personal data is being used or to consent to such use.
EPIC further contends HireVue's practices are inconsistent with public policy, as set forth by the Organization for Economic Cooperation and Development's Principles on Artificial Intelligence and the Universal Guidelines for Artificial Intelligence, two sets of international guidelines that create a framework for governance of artificial intelligence but which have not been codified in the United States.
EPIC urges the FTC to open an investigation into HireVue's practices, issue an injunction, and require HireVue to: (1) make public the algorithm used to produce assessments of job candidates, and (2) make known to candidates the precise basis for their evaluations.
It is unclear at this early stage what the FTC will do with EPIC's complaint. At a minimum, the EPIC complaint publicly develops some of the arguments consumers and employees might advance against automated decision-making, provides another potential path by which individuals might challenge such processes, and reflects the FTC's growing role in regulating the technology.