The U.S. Department of Justice recently penalized a Virginia-based company for using AI job postings that unlawfully excluded American workers. While the $9,460 fine against Elegant Enterprise-Wide Solutions may appear nominal, the settlement brings a shift in enforcement. Regulators are keen to place the full burden on employees who deploy automated tools for simple tasks such as job postings.
The DOJ’s Civil Rights Division found that the Virginia firm used an AI tool to not only draft but also disseminate its job postings. These postings explicitly restricted applicants to specific visa categories, such as H-1B and OPT, effectively screening out U.S. citizens and green card holders.
Employers are now required to display federal “Right to Work” notices in all applicant-facing locations.
Unfortunately for Elegant Enterprise-Wide Solutions, this practice violates that Immigration and Nationality Act (INA), which prohibits citizenship-status discrimination. Assistant Attorney Kristen Clarke was unequivocal on this part. The Department has made a clear stance that it will not tolerate discrimination, regardless of whether the recruiter is a human or an AI.
What does this mean for companies outsourcing their HR to AI? The answer is clear. Do not outsource your legal liability.
Can AI job postings be held legally liable for discrimination?
This is the tricky part. While Elegant Enterprise Wide Solutions case focused on the employer, there are other high-profile litigations in similar hot waters.
The current federal stance, supported by the EEOC and DOJ, is that automated systems must follow the same civil rights standards as traditional hiring. If an algorithm is trained on biased data or programmed with discriminatory filters, the resulting disparate impact is a direct violation of federal law.
For the Virginia firm, the settlement mandates a three-year monitoring period, during which the DOJ has the authority to inspect premises and audit internal records.
What does this mean for employers?
For starters, all HR staff must undergo intensive training on INA non-discrimination provisions. In addition, the company must revise all recruitment protocols and have them approved by the DOJ. Employers are also required to display federal “Right to Work” notices in all applicant-facing locations.
How to guard yourself against AI bias?
It’s time for HR leaders to treat AI as a collaborator, not a substitute. This means companies must undergo ‘bias audits’ and ensure that human-in-the-loop protocols are the basic requirement.
As we move closer towards the end of the decade, the so-called black box of AI is being cracked open by federal investigators. Companies that fail to audit their AI job postings may find themselves in hot waters tomorrow.
The DOJ’s enforcement action shows that employers are strictly liable for AI-generated content. In order to reduce the risk of bias and fines, companies must implement mandatory human reviews of all AI-produced job ads and conduct regular audits of their recruitment software.
Subscribe to The HR Digest for more insights on workplace trends, layoffs, and what to expect with the advent of AI.
