Meta Wins Initial Court Ruling As Judge Refuses To Halt AI-Linked Layoffs Amid Novel Discrimination Challenge

A U.S. federal judge has refused to temporarily block Meta Platforms from proceeding with layoffs affecting 26 employees who claim the company used artificial intelligence tools in a way that discriminated against workers with disabilities or those who took medical or family leave.

The ruling allows Meta to move forward with the layoffs beginning July 22 while the employees pursue their claims through private arbitration, marking an early victory for the social media giant in what appears to be one of the first major legal challenges in the United States over the alleged use of AI in workforce reduction decisions.

U.S. District Judge William Orrick in Oakland, California, ruled that the workers failed to demonstrate the “irreparable harm” necessary to justify an emergency order preventing the layoffs before the underlying dispute is resolved. The decision addresses only the request for immediate relief and does not determine whether Meta’s use of AI in its employment decisions violated anti-discrimination laws.

Meta has denied wrongdoing, maintaining that human managers, not AI systems, made the final layoff decisions.

The lawsuit was filed after Meta informed nearly 8,000 employees in May, representing about 10% of its global workforce, that their positions would be eliminated as the company reallocates resources toward artificial intelligence development and infrastructure. The workforce reduction forms part of CEO Mark Zuckerberg’s broader strategy to accelerate Meta’s AI ambitions, including heavy investments in AI models, computing infrastructure and talent.

The plaintiffs, who filed anonymously, include engineers, managers, researchers, and designers. Although they lost access to Meta’s internal systems in May, they have remained on the company’s payroll pending the completion of the layoffs, with many terminations scheduled for July 22 and others later in July or August.

The workers argue that Meta relied on AI-powered evaluation systems that measured productivity, AI token usage and employee adoption of AI tools when identifying staff for dismissal. According to the complaint, these systems allegedly disadvantaged workers who had taken legally protected medical or family leave because their productivity metrics and AI usage naturally declined during their absence.

The lawsuit identifies several internal systems allegedly used in the process, including an AI assistant known as “Metamate,” an employee-trained knowledge platform described as a “second brain” that tracked workplace communications and documents, and productivity scores reportedly generated by monitoring keystrokes, screen activity, emails and browser history.

The plaintiffs allege that these AI-driven metrics continued operating while employees were on leave, reducing their performance rankings and increasing their likelihood of being selected for layoffs.

Meta disputes those allegations, insisting that AI tools did not independently determine who would lose their jobs and that human decision-makers remained responsible for the final selections.

During Thursday’s court hearing, lawyers for the employees argued that immediate layoffs would cause lasting harm beyond lost wages. They said workers faced the loss of valuable stock compensation, employer-sponsored health insurance, and medical coverage for pregnancies and ongoing treatments.

“There’s no do-over for bonding with a new baby or giving birth or having active medical treatment,” attorney Barbara Cowan told the court.

Meta countered that employees would lose only employer-subsidized insurance rather than healthcare coverage entirely, arguing that any financial losses could be compensated if the workers ultimately prevail in arbitration.

Judge Orrick agreed that the alleged harm did not meet the high legal standard required for emergency injunctive relief. However, he indicated he is likely to rule next month on the workers’ request for a preliminary injunction, which would provide longer-term temporary relief if granted.

The case also highlights the growing legal scrutiny surrounding the use of artificial intelligence in employment decisions. While companies increasingly deploy AI to assist with hiring, performance management and workforce planning, regulators and employment lawyers have warned that algorithmic systems can inadvertently disadvantage protected groups if the underlying data or evaluation methods are biased.

The lawsuit could become a landmark test of how existing employment discrimination laws apply to AI-assisted workplace decisions, particularly where algorithms influence productivity assessments or layoff recommendations.

Another notable aspect of the dispute involves mandatory arbitration agreements, which have become common across major U.S. technology companies. Meta requires employees to resolve workplace disputes through individual arbitration rather than class-action litigation.

The plaintiffs argue that while their employment agreements require arbitration of the underlying claims, they do not prevent workers from seeking temporary court intervention to stop allegedly unlawful employment actions before arbitration concludes. Such requests for emergency relief are more commonly associated with trade secret disputes or cases involving departing employees accused of soliciting customers, making this attempt to halt AI-assisted layoffs relatively unusual.

The case comes as employers across industries increasingly incorporate AI into human resources functions. Governments in the United States and Europe have responded with closer oversight, reflecting concerns that automated decision-making systems may replicate or amplify existing workplace biases if not properly monitored.

Although Friday’s ruling allows Meta to proceed with the layoffs, the broader legal questions surrounding the company’s alleged use of AI in employment decisions remain unresolved and could help shape future standards for AI governance in the workplace.