26 Meta Workers Take Algorithmic Job Cuts to Court in 2026

Twenty-six current and former Meta employees have taken the company to federal court, and the heart of their case is a question every worker should be asking in 2026. Can an employer let software decide who loses their job? As CBS News and Fortune reported, the workers filed suit in federal court in Oakland, California, claiming Meta used a stack of internal artificial intelligence systems to rank employees and help select roughly 8,000 people for layoffs, about 10 percent of its workforce. The workers say those algorithms counted their protected medical, pregnancy, and family leave against them as low performance. At Metaintro, we track how AI is reshaping who gets hired and who gets cut across the labor market, and this case sits right on that fault line. Here is what the lawsuit actually claims and what it means for your career.

What exactly are the 26 Meta workers accusing the company of?

The complaint, filed late Monday in federal court in Oakland, describes something more specific than a routine round of layoffs. According to CBS News, the workers say Meta leaned on a set of internal AI systems, keystroke and activity-monitoring data, AI token-usage dashboards, and algorithmically assisted performance rankings to help decide who would be laid off. In plain terms, the software watched how much people typed, how often they were active, how much they used internal AI tools, and how a model scored their output, and then it helped sort employees into keep and cut piles.

The plaintiffs argue that this design has a built-in flaw. Many of those scores and ratings, the lawsuit says, "by design, cannot be accumulated by an employee who is on protected medical or family leave, or whose output is reduced by a disability." If you are out on approved leave, you are not typing, not logging activity, and not producing measurable output, so the system reads your absence as poor performance. The suit claims Meta "did not pause the system for the individualized, leave- and accommodation-neutral review that the law requires." Meta, for its part, told reporters the claims "lack merit and are not based on facts" and that "workforce management and organizational decisions were and are made by people, not AI." That single sentence, whether people or machines made the final call, is likely to become the central fight of the case. This is the same tension we covered when bosses pushed workers to use AI and it backfired, and it is why we keep telling readers at Metaintro to understand exactly how their employer measures them.

Who got caught in Meta's algorithmic layoffs?

The 26 plaintiffs are anonymous in the filing, but the pattern in their stories is not subtle. According to Fortune, each of the 26 had taken protected leave and had requested or received a reasonable accommodation for a disability. About half of them took leave for caregiving or pregnancy-related reasons. Eight are women who had taken maternity or pregnancy-related leave, four are men who had taken parental leave, and one is a woman who took time off to care for a family member and later took bereavement leave.

One account stands out. The lawsuit describes an employee who disclosed a serious health condition and disability that was approved by Meta's own medical provider, then says he was "discouraged and deterred from taking that leave by a manager" who warned that using it would land him on the layoff list. Meta, the suit says, offered no accommodation for his disability. If true, that is exactly the kind of story that turns a data dispute into a human one, and it is the reason worker-protection cases can gain traction even in a tough legal climate. Being flagged by a system you cannot see is disorienting, and workers who have been through it describe the same scramble we detailed in the story of an Atlassian engineer who fought to win her job back after a layoff. The emotional weight is real, and so is the financial one, especially when a cut lands during pregnancy or active medical treatment.

How can an algorithm end up discriminating against people on leave?

This is where the case gets bigger than Meta. The legal theory driving it is called disparate impact, a longstanding civil rights idea explained by CBS News and codified in Title VII of the 1964 Civil Rights Act. Disparate impact holds that a policy can be illegal even if it looks neutral on its face, as long as it disproportionately burdens a protected group and is not necessary for the job. A performance score does not say "penalize pregnant employees." It simply measures output. But if output naturally drops to zero during protected leave, the neutral-looking score quietly lands hardest on the people the law is supposed to shield.

The plaintiffs' lawyers put it directly, arguing that the "algorithmically assisted selection process, by systematically recording such absences as reduced performance, falls more heavily on women than on men," because women disproportionately take pregnancy and caregiving leave. That is the whole problem with outsourcing judgment to a model that only knows activity data. It cannot tell the difference between someone who checked out and someone who is legally, and temporarily, absent. It also cannot see context that any human manager would weigh, the parent who shipped a quarter of brilliant work before going on maternity leave, or the employee who negotiated a disability accommodation in good faith. When a score treats a protected absence as a gap in output, the bias is not a bug that slipped through. It is the predictable result of measuring the wrong thing and trusting the number anyway. That is a risk for early-career workers too, who often have the thinnest track record for a model to read, a pressure we explored in our reporting on young adults and the shrinking pool of entry-level jobs. We have watched versions of this play out across the labor market, from the rise of new Luddite workers pushing back on AI to companies quietly rehiring the workers they replaced with AI once the automation fell short. The lesson for job seekers reading Metaintro is that the algorithm deciding is never a neutral outcome. It reflects whatever the people who built it chose to measure.

Which laws does the lawsuit say Meta broke?

The complaint stacks up an unusually long list of protections. Per CBS News and Fortune, the workers say the layoffs violated the Family and Medical Leave Act, the Americans with Disabilities Act, the Pregnancy Discrimination Act, and the Pregnant Workers Fairness Act. Together those laws guarantee unpaid job-protected leave, require reasonable accommodations for disability and pregnancy, and bar employers from treating protected time off as a mark against you.

There is a political wrinkle worth understanding. The complaint leans on disparate impact liability at a moment when the current administration has moved to abandon it, ordering federal agencies to deprioritize enforcement and prompting the Equal Employment Opportunity Commission to drop some discrimination cases. But as CBS News notes, workers can still bring these claims themselves when the EEOC steps back, and several state laws specifically prohibit disparate impact discrimination. The suit also cites a landmark 1971 Supreme Court ruling that first recognized the doctrine. For workers, the practical takeaway is that your rights around leave and accommodation do not disappear because enforcement priorities shifted, a point we made when breaking down a recent PIP court ruling and what workers should know. Knowing which statute protects you is the difference between accepting a cut and challenging one.

Why does this case matter even if you do not work at Meta?

Because the tools at the center of it are already everywhere. Activity monitoring, productivity scoring, and AI-assisted performance ranking are no longer exotic. They increasingly sit inside ordinary HR software at ordinary employers, which means the exact mechanism the Meta plaintiffs are challenging could be running quietly in workplaces far smaller than a tech giant. We have tracked the wider wave in pieces on Allianz cutting jobs as it leans into AI adoption, on Citigroup restructuring under Jane Fraser, and on the Philippines BPO sector shedding call-center roles to AI.

A ruling in the workers' favor, or even a high-profile settlement, could push every employer using these systems to audit them for bias before the next round of cuts. That is not a far-fetched hope. The lawsuit itself points out that companies remain exposed to disparate impact claims in the age of AI, and it notes that new laws in some jurisdictions already require testing automated decision tools for bias. Meanwhile the broader numbers keep climbing, as we documented in our look at the United States leading the world in 2026 tech layoffs by the numbers and in the ripple effects of Amazon layoffs crushing workers in a saturated job market. If a machine can help decide your job, you deserve to know how it scores you, and that is a fight the Meta case may help every worker win. It is also why Metaintro exists, to give job seekers a clearer view of the market than their employer's dashboard gives them.

What does this mean for your career and your rights?

Start by treating your own performance record as something you control, not something a dashboard owns. Keep your own log of shipped work, wins, and manager praise, because an AI score built on activity data will never capture the weeks you were legally on leave. If you take or plan to take medical, pregnancy, or family leave, get the approval in writing and understand that neither the Family and Medical Leave Act nor the Americans with Disabilities Act allows that time to be counted against you. If you sense a layoff coming, learn what your package should include before you sign anything, the way we walked through Microsoft's unusually generous layoff severance and in our guide to how to negotiate a job offer.

It also helps to understand the culture you are being scored inside of, because the same companies that lean hardest on monitoring dashboards are often the ones where workers feel least safe raising a concern, a dynamic we unpacked in our piece on why so many workers will not bring a toxic workplace problem to HR. If your performance is being tracked by software, ask for the criteria in writing, ask how leave is handled, and save the answer. A paper trail is the single most useful thing you can build, both for a legal claim and for your own peace of mind.

Then build your exit ramp before you need it. Workers who are clinging to their roles out of fear, a trend we covered in the great hunkering down, are often the least prepared when the cut finally comes. The healthier move is to keep your options warm, benchmark your value with tools like our guide to pay transparency laws and what you can ask, and have a search plan ready. If a layoff does land, a structured restart beats panic, which is exactly why we built our playbook on how to run a job search with AI after a layoff. The Meta workers are fighting to stay employed, but their real message to the rest of us is simpler. Understand how you are being measured, know the rights that protect you, and never let a single algorithm be the only thing standing between you and your next paycheck. At Metaintro, that is the whole point.

Related Articles

• The United States is leading the world in 2026 tech layoffs by the numbers
• Amazon layoffs are crushing workers in a saturated job market
• An Atlassian engineer won her job back after a layoff fight
• What workers should know about a recent PIP court ruling
• How Microsoft's layoff severance stacks up for workers
• Allianz is cutting jobs as AI adoption accelerates
• Companies are rehiring the workers they replaced with AI
• The new Luddites are workers pushing back on AI
• How to run a job search with AI after a layoff
• The great hunkering down as workers cling to jobs and perks
• Pay transparency laws and what you can ask

People Also Asked

Q: Did Meta really use AI to decide who got laid off?

A: The 26 workers claim Meta used internal AI systems, activity and keystroke monitoring, token-usage dashboards, and algorithmic performance rankings to help select employees for its roughly 8,000 layoffs, as reported by CBS News. Meta denies it, saying those decisions "were and are made by people, not AI." A court has not ruled on the facts yet, so the question of how much the software actually decided is still open.

Q: Is it legal for a company to lay you off while you are on medical or family leave?

A: Not if the leave itself is why you were selected. The lawsuit argues Meta violated the Family and Medical Leave Act, the Americans with Disabilities Act, and pregnancy protections by treating protected absence as low performance. These laws generally bar employers from counting approved leave against you, though the details depend on your situation and your state.

Q: What should I do if I think an algorithm hurt my performance review?

A: Keep your own written record of your work and any approved leave, ask your employer in writing how performance is measured, and confirm what accommodations you requested and received. If you believe a cut was discriminatory, you can file a complaint with the Equal Employment Opportunity Commission or consult an employment lawyer, and you can start building a backup plan with a structured job search after a layoff.

Layoffs are hard, and an algorithm that scored you in the dark makes them harder, but you do not have to face the next move alone. Metaintro helps you see where the real opportunities are, benchmark what you are worth, and line up your next role before you need it. Looking for your next opportunity? Create your free Metaintro profile and put the market to work for you.