Twenty-six Meta employees say the company turned its own artificial intelligence tools into a hidden layoff machine that hit people hardest when they stepped away to have a baby, fight illness, or care for family.
Story Snapshot
- Twenty-six current and former Meta workers filed a federal lawsuit over 2026 layoffs.
- The suit claims Meta’s internal artificial intelligence systems helped pick who got cut, punishing workers on protected leave.
- Plaintiffs say activity scores, keystroke tracking, and “second brain” tools cannot be earned while on medical or parental leave.
- Meta denies the claims, saying people, not artificial intelligence, made the layoff decisions.
A lawsuit that turns Meta’s own tools into Exhibit A
A group of twenty-six Meta employees went to federal court in Oakland, California, saying the company’s high-tech management tools did something very old-fashioned: punished people for taking care of their health and their families.
The case centers on mass layoffs that cut about eight thousand jobs, roughly ten percent of Meta’s workforce, after workers were told in May their positions would be gone starting July twenty-second. These workers say the way Meta chose who stayed and who left crossed legal lines.
The lawsuit says Meta did not just rely on managers looking at performance reviews. Instead, it used internal artificial intelligence-assisted systems, activity monitoring, and algorithmic rankings to build a termination list.
These systems allegedly scanned keystrokes, screen content, emails, browser history, code output, and how much workers used Meta’s own artificial intelligence tools.
On paper, it sounds like a hyper-efficient business dashboard. In practice, the plaintiffs say, it turned into a quiet trap for anyone whose work slowed down because they had a baby, surgery, disability, or family crisis.
How activity scores collide with protected leave
The workers’ core claim is simple enough to make sense to any parent or patient: you cannot rack up keyboard hits or artificial intelligence “token” usage when you are not supposed to be working.
The complaint says many of the scores Meta used “by design, cannot be accumulated by an employee who is on protected medical or family leave, or whose output is reduced by a disability.”
That means the very metrics Meta trusted will always make a person on approved leave look slow or absent when compared to a coworker who never stepped away.
A lawsuit accusing Meta of discriminatory artificial intelligence use in firing workers demonstrates how companies leaning on the emerging technology for employment decisions can incur liability. https://t.co/jpeh9SPay7
— Bloomberg Law (@BLaw) July 28, 2026
About half the plaintiffs had taken leave for caregiving or pregnancy-related reasons. Eight are women who took maternity or pregnancy leave, four are men who took parental leave, and one woman took family care and later bereavement leave.
Others reported medical conditions and disabilities that reduced their output. The suit even describes a worker who received a layoff notice while on approved pre-birth leave, just two days before giving birth.
This looks less like “hard-nosed business” and more like using a fancy dashboard to dodge basic fairness toward people who did exactly what the law encourages: take care of their families without hiding it.
The legal lines: when scoring becomes discrimination
The lawsuit does not just complain about hurt feelings or rough corporate culture. It points straight at federal and state laws that protect workers who take medical, parental, or caregiving leave, or who live with disabilities.
The filing cites the Family and Medical Leave Act, the Americans with Disabilities Act, the Pregnancy Discrimination Act, and the Pregnant Workers Fairness Act, among others.
Those laws are not niche technical rules. They reflect a basic American idea: you should not lose your livelihood because you took lawful time to heal, give birth, or care for a loved one.
The plaintiffs say Meta broke that promise by failing to adjust or “pause” its systems for leave takers. Instead of an individualized review that treats approved absences as neutral, they allege Meta let the machines run and simply counted everyone’s activity as if no one ever has a baby or chemo appointment.
From a rule-of-law standpoint, this is the heart of the dispute. If a company builds a scoring engine it knows will tank the numbers of people on protected leave, then still uses those numbers to pick who gets fired, it starts to look like discrimination dressed up in data.
Meta’s denial and the new proof problem of artificial intelligence layoffs
Meta has pushed back hard in public. A company spokesperson said the claims “lack merit and are not based on facts,” and insisted that “workforce management and organizational decisions were and are made by people, not artificial intelligence.”
That statement draws a bright line, but it raises a sharp question: if people leaned on AI-powered rankings and dashboards, where does the machine end and human judgment begin? For workers trying to protect their rights, that line matters less than the outcome: who got cut, and why.
The workers asked the court for emergency relief, hoping to block Meta from finishing the layoffs while their individual claims move through private arbitration.
The judge declined to halt the cuts, stressing that the employees “were not in the rooms where it happened,” which highlights the new proof problem in the age of artificial intelligence management.
Meta holds the logs, models, and dashboards that could show whether people on leave were treated differently. Until those records surface, the public sees a familiar clash: workers describing patterns of harm, and a giant company demanding trust in its internal process without showing its math.
Sources:
abc7.com, theguardian.com, reuters.com, youtube.com
