Meta used AI tools to pick 8,000 workers for layoffs, lawsuit claims
Twenty-six former employees say an algorithm scored and ranked them for termination, and that the system penalised workers who took medical leave or had disabilities.
Key points
- A lawsuit filed by 26 unnamed plaintiffs on 17 July 2025 claims Meta used internal AI systems to build its layoff list, not human managers.
- Meta's internal dashboards reportedly sorted employees into categories including "AI Native", "AI First", and "AI Enabled" based on how much they used AI tools.
- The complaint, filed in the US District Court for the Northern District of California, says workers with disabilities and those on protected medical or family leave were disproportionately selected.
- One system named in the lawsuit is called "Metamate", an internal AI tool Meta used to monitor and score employee activity.
Meta cut 8,000 jobs earlier this year. Most big layoffs follow a familiar script: senior managers review teams, weigh performance, and make hard calls. This lawsuit says Meta skipped that part.
According to the complaint, first reported by Ars Technica, Meta assembled its termination list using a collection of internal AI systems. One is called "Metamate". Another is described as a "second-brain" agent, meaning software trained on employee data that can carry out analysis on its own. The tools also pulled in keystroke and activity logs, dashboards tracking how often workers used AI products, and an algorithmic performance-ranking system.
In plain terms: a set of automated tools watched what employees did, scored them, and surfaced names for the cutting list.
One scoring factor, the lawsuit alleges, was how enthusiastically employees adopted Meta's own AI tools. Workers were reportedly labelled by three tiers: "AI Native" at the top, then "AI First", then "AI Enabled". If you were not using Meta's internal AI products enough, the system counted that against you.
Did the AI unfairly target protected groups?
Yes, that is the core legal claim. The 26 plaintiffs say workers with disabilities and those who had recently taken protected medical or family leave scored lower under the automated system, because illness and recovery time naturally reduce the kind of digital activity the tools were measuring. A person on cancer treatment cannot rack up AI-tool usage hours.
Federal law, specifically the Americans with Disabilities Act and the Family and Medical Leave Act, bars employers from punishing workers for those exact circumstances. The lawsuit argues Meta's automated scoring did exactly that, just without a human manager ever consciously deciding it.
Meta has not yet responded publicly to the allegations, and a court has not ruled on the merits.
What this means for workers everywhere
This case matters beyond Meta. Companies across every industry are quietly adopting productivity-monitoring software and AI scoring tools. If you are an employee, it is worth asking your HR department whether automated systems play any role in performance reviews or staffing decisions. That is not paranoia; it is the kind of question that has a factual answer.
The honest takeaway: if you take legally protected leave and later face a poor performance review or job loss, document everything. Dates, communications, any metrics you were shown. A paper trail is your clearest protection if automated scoring is ever used against you.
