Employment Law Update: A Scheduling System Cost One Employer $125 Million. Now 26 Employees Are Suing Over AI-Driven Layoffs. Is Your Organization Next?
What Employers Need to Know About AI Potential Liability in Hiring, Scheduling, and Workforce Reduction Decisions
Date: July 24, 2026
By:
Lisa M. Brauner
The jury deliberated for three hours. It returned a $125 million punitive damages verdict (the court subsequently reduced the punitive damages to $300,000, the ADA’s statutory cap for large employers, but also awarded reinstatement, backpay, prejudgment interest, and other relief). The Seventh Circuit affirmed.
The scheduling system was not AI by today’s standards. Humans made every decision that followed; they simply chose not to act. The takeaway? The ADA’s interactive process obligation does not come with a carve-out for automation.
In 2026, the risk has escalated. This month, 26 current and former Meta Platforms, Inc. (“Meta”) employees filed what appears to be the first major lawsuit challenging AI-driven layoff decisions. The complaint alleges that, in May 2026, Meta began notifying approximately ten percent of its workforce of their selection for termination. The plaintiffs allege Meta used a constellation of internal AI systems — keystroke and activity-monitoring data, AI token-usage dashboards, and algorithmically assisted performance ranking- to score, rank, and select employees for the termination list. Those systems draw on inputs that an employee on protected leave cannot accumulate: productivity metrics, “AI-native” ratings, and AI-token consumption.
The complaint alleges Meta did not neutralize those inputs for protected leave, did not exclude leave-takers from the scoring cohort, and did not pause for the individualized review that federal and state law requires. The plaintiffs claim that the system recorded protected-leave time and disability-related output reductions as underperformance, producing a disparate impact on the basis of sex, pregnancy, and disability, and that Meta never tested the process for bias. One plaintiff, a scientist on approved pregnancy leave, was notified of her selection for termination two days before giving birth.
The plaintiffs assert claims for both disparate treatment and disparate impact. They allege the algorithmically-assisted selection process discriminated on the basis of sex, pregnancy, and disability, and used protected leave and disability accommodations as a negative factor in selecting employees for termination. The central questions: (1) whether Meta used protected-leave status, accommodation status, or proxies for either as a negative factor; (2) whether Meta tested the process for disparate impact; and (3) whether the process produced an unjustified disparate impact. The complaint invokes the Family and Medical Leave Act, Title VII as amended by the Pregnancy Discrimination Act, the Americans with Disabilities Act, the Pregnant Workers Fairness Act, and state anti-discrimination laws.
Any system that measures productivity without adjusting for protected leave is not a neutral tool; it is a liability engine running on your data.
The takeaways for employers:
- Audit every AI system that touches employment decisions. Assess hiring, scheduling, performance scoring, and layoff tools for disparate impact on employees with disabilities or on protected leave.
- Ask whether your system treats protected leave as underperformance. If a productivity score drops because the system records leave time as a gap rather than excluding it, the system is converting a legal right into a negative performance signal, supporting potentially both disparate-impact and disparate-treatment liability.
- Build accommodation checkpoints into the workflow. Scores must be adjusted or excluded for anyone exercising ADA, FMLA, or state or city law-protected rights.
- Never skip the interactive process. No algorithm replaces the employer’s obligation to engage individually with an employee who has a known disability or to know who is on protected leave.
- Test for bias before you deploy. Some jurisdictions like California, Illinois, and New York City prohibit employers from using automated decision systems or criteria that result in discrimination against protected categories. NYC requires employers to have an independent auditor conduct an annual bias audit before using AI or automated tools to screen job candidates. Texas prohibits employers from developing or using AI systems to intentionally discriminate against a protected class in violation of federal anti-discrimination laws. Other jurisdictions are following, in various respects.
- Monitor AI outputs on an ongoing basis. A pre-deployment audit is not enough. Review AI-generated scores quarterly, disaggregated by disability status, leave status, and other protected characteristics.
- Negotiate AI-specific indemnification in vendor contracts. Require representations from your vendors that the tool has been tested for adverse impact, obligations to disclose algorithm changes, indemnification for discriminatory outputs, and audit cooperation. An employer may remain liable regardless of who built the system; indemnification shifts the financial exposure.
- Ensure meaningful human oversight. No AI tool should produce a termination, discipline, or layoff recommendation without a trained human reviewing the output for legal compliance before it becomes final.
In Spaeth, a basic scheduling system changed one employee’s shift by 90 minutes. The employer refused to accommodate. A jury returned a nine-figure verdict in three hours. In the Meta case, the allegations are that AI systems scored an entire workforce on metrics that employees on protected leave could not accumulate, and that no one paused to ask whether the process was lawful before 26 employees were selected for termination. The legal principle is the same: automation does not excuse the employer’s obligation to account for protected status. The scale of exposure may not be the same in both cases. Act now; not after the next jury verdict.
ABOUT OUR TEAM
Whiteford's Labor & Employment Group represents management in all aspects of the employment relationship, from proactive compliance counseling and legal audits of workplace practices to employee training to internal investigations to defending discrimination, accommodation and retaliation claims in federal and state courts. If your organization is deploying AI or automated tools in hiring, performance management, or workforce reduction decisions, we can help advise with respect to audits of those systems, help build legally defensible accommodation processes, and stay ahead of rapidly evolving federal, state and local law requirements.Lisa M. Brauner is a partner at Whiteford in its Labor & Employment Section in New York and has been recognized as a New York Super Lawyer and named among Best Lawyers in America.® Lisa can be reached at Lbrauner@whitefordlaw.com and 646-618-8655.
The information contained here is not intended to provide legal advice or opinion and should not be acted upon without consulting an attorney. Counsel should not be selected based on advertising materials, and we recommend that you conduct further investigation when seeking legal representation.