A group of 26 former Meta employees has sued Meta, alleging the company used internal AI systems to help choose workers for layoffs in a way that unfairly hit people on parental or medical leave, according to Reuters and the employees’ lawsuit.
The allegation is not that Meta laid people off with a single magic robot button. The former employees say Meta fed performance information into what the complaint calls a “constellation” of internal AI tools, then used those outputs to rank employees for dismissal. Their central claim is that Meta did not properly remove or adjust for workers who had taken protected leave before applying that ranking.
According to the lawsuit, that design meant employees who used parental or medical leave were more likely to be selected for layoffs. The complaint says the scoring system failed to account for time away from work and, in practice, punished employees for using leave the lawsuit describes as protected.
That mechanism matters. Performance systems often depend on signals such as recent work output, manager feedback, project participation, and other workplace activity. If a ranking system treats an employee’s reduced activity during leave as a performance problem, rather than as an absence that should be excluded or adjusted for, the score can make a legally protected absence look like underperformance. The former workers allege Meta’s tools did exactly that.
The lawsuit places AI in the middle of an old employment problem: how companies convert messy personnel records into layoff lists. Meta is accused of using automated or AI-assisted analysis without adequate guardrails for people whose records reflected parental or medical leave. The complaint does not claim the tools were neutral and then misread by managers. It alleges the ranking process itself failed to protect leave-taking employees from being disadvantaged.
Reuters reported the lawsuit earlier. The available allegations do not establish that a court has found Meta liable, and Meta’s response is not included in the complaint materials described here. The case is at the accusation stage, where the former workers must prove that Meta’s systems and decision-makers caused the claimed disparity.
The dispute adds a sharper edge to corporate use of AI in human resources. Companies like Meta can call these systems tools, rankings, or internal analytics, but the lawsuit’s point is blunt: if the output helps decide who loses a job, the design choices behind the score are employment decisions too.
This story draws on original reporting from The Verge AI.
