This article was originally published on Crafting Your Home. A human contributor also wrote and edited the post.
Meta is facing a potentially landmark legal battle after 26 employees accused the technology giant of using artificial intelligence systems to help decide who would lose their jobs during a sweeping round of layoffs. The federal lawsuit alleges that workers who took maternity, medical, or parental leave, or received disability accommodations, were pushed toward termination because automated performance tools treated protected absences as poor productivity.
The employees are asking a judge in Oakland, California, to stop their terminations before they begin on July 22, 2026. They were notified in May that their jobs would be eliminated as part of Meta’s plan to cut roughly 8,000 positions, about 10 percent of its global workforce. Their case raises a larger question: what happens when machines built to measure performance cannot understand pregnancy, illness, caregiving, or disability?
Meta denies the allegations and says people, not AI, made the decisions. The court has not ruled on the claims.
Workers Say Meta’s AI Systems Turned Protected Leave Into Poor Performance
The workers allege Meta used internal AI-assisted tools to score, rank, and identify people for termination, rather than relying solely on managers familiar with their work. The systems allegedly included Meta’s internal assistant, Metamate; employee-trained “second brain” agents; AI token-usage dashboards; and productivity measurements drawn from digital activity.
According to the lawsuit, the tools considered signals such as output, AI use, keystrokes, screen activity, emails, and browser history. That appears precise until an employee disappears from the dashboard for a legally protected reason. Someone recovering from surgery, caring for a newborn, or managing a disability will naturally produce fewer clicks and fewer completed tasks.
The plaintiffs say the systems recorded that absence as reduced performance without properly accounting for why the person was away. Every plaintiff had taken protected leave or requested or received a disability accommodation. About half had taken leave linked to pregnancy, parenting, or caregiving.
The group includes eight women who took maternity or pregnancy-related leave, four men who took parental leave, and one woman who took caregiving and bereavement leave. The individual accounts make the allegations especially striking. One scientist on approved pre-birth leave reportedly learned she was being laid off two days before giving birth.
An engineer claimed his rating dropped after time away for an injury. A manager on medical leave was allegedly selected for termination 16 days into his absence. Another worker said a manager warned that taking approved medical leave could increase his risk of being cut. The employees say Meta failed to adjust its scoring for individualized review, leaving workers on protected leave unable to earn the same digital scores as active colleagues.
Meta’s Monitoring Program Fuels Questions About Privacy and Bias
The lawsuit also puts Meta’s employee-monitoring practices under scrutiny. The company introduced an AI monitoring program earlier in 2026 designed to capture activity on company devices, including keystrokes, mouse movement, browser history, messages, emails and location information. The goal was reportedly to train AI systems by observing how employees worked.
That ambition creates an uncomfortable contradiction. Meta has presented AI as a tool that can make organizations faster and leaner. Yet the lawsuit suggests the same technology may have judged workers through a narrow digital lens, rewarding constant, measurable activity while overlooking the reasons for an absence. A dashboard can count messages.
It cannot understand a difficult pregnancy, serious injury, or caregiving crisis unless those protections are built into the system. The complaint alleges that the program arrived with little transparency and limited consent. Some teams reportedly received no clear acknowledgment prompt, and workers initially had no practical way to opt out. More than 1,600 employees later signed a privacy petition, and CEO Mark Zuckerberg announced a pause in June after the backlash.
Automated employment systems are only as fair as their data, assumptions, and human review. If productivity is defined through keystrokes, AI prompts, and visible activity, employees with disabilities or protected absences may start every comparison at a disadvantage. A system can look neutral while repeatedly producing unequal results.
The plaintiffs want an independent audit and a clearer explanation of how they were selected. Their lawyers say Meta kept the process secret. Without knowing what information was used, how it was weighted, and whether managers corrected distortions, employees have little chance to challenge the result.
Meta says humans made the decisions. The lawsuit may now force the company to explain how heavily those humans relied on automated rankings, because final human approval does not settle whether technology shaped the list.
The Lawsuit Could Redefine Accountability for AI-Driven Layoffs
The plaintiffs allege that Meta violated the Family and Medical Leave Act, the Americans with Disabilities Act, the Pregnancy Discrimination Act, and the Pregnant Workers Fairness Act. They also allege that Meta failed to test its systems for bias under newer rules governing automated workplace decisions.
A key argument involves “disparate impact,” the principle that a neutral-looking policy may still be discriminatory if it burdens a protected group without sufficient job-related justification. The workers say continuous-activity scoring may fall more heavily on women because women disproportionately take pregnancy and caregiving leave. Although the Trump administration has reduced federal emphasis on disparate-impact enforcement, private workers may still bring claims, and several states maintain protections.
Apparently, the first lawsuit against a major American company challenges the alleged use of AI in mass layoffs. The dispute therefore reaches beyond Meta. Technology companies are investing heavily in artificial intelligence while cutting staff and asking smaller teams to produce more. Meta eliminated about 10 percent of its workforce in May while expanding costly AI programs and pursuing greater efficiency.
For the 26 employees, the stakes are immediate. Their lawyers say separation could end employer-funded health coverage during pregnancy, postpartum recovery, or medical treatment. It could also erase leave rights, cost workers unvested equity, and trigger immigration problems for employees whose legal status depends on their jobs. They want the court to preserve their employment while individual claims proceed through arbitration.
The broader stakes reach every workplace adopting automated management. AI can process enormous amounts of information, but speed does not guarantee fairness. A system that cannot distinguish low effort from protected absence may scale discrimination with remarkable efficiency. A manager who accepts that output without careful review does not restore human judgment.
The manager simply gives the algorithm a signature. Meta may ultimately defeat the claims, and the employees must still prove that AI played the role they describe.
Yet the lawsuit has exposed the danger of allowing opaque technology to influence employment decisions without strong safeguards, independent testing and meaningful oversight. The dispute concerns 26 workers, but its outcome could shape how millions of employees are measured, monitored and dismissed in the age of artificial intelligence.
If you like what you just read, then subscribe to our newsletter and follow us on social media
