A group of 26 Meta employees has filed a lawsuit against the company, alleging that it employed artificial intelligence systems to determine which employees would be laid off. The lawsuit claims this method disproportionately affected those on medical, parental, or family leave. In May, Meta announced that it would lay off around 8,000 employees, which represents about 10% of its workforce.
The lawsuit, filed in federal court in Oakland, California, contends that Meta relied on various internal AI systems, including keystroke and activity-monitoring data, AI token-usage dashboards, and algorithmically assisted performance rankings, to decide who would be terminated. The plaintiffs argue that these scoring systems were inherently biased against employees taking protected leave, as their productivity metrics were adversely impacted during their time off.
According to the complaint, Meta failed to consider the implications of protected leave when calculating employee scores and did not implement the individualized reviews that the law mandates. This disregard resulted in employees on medical or family leave being more likely to face layoffs. All 26 anonymous complainants in this case had taken protected leave and had requested or received reasonable accommodations for disabilities. While they have been informed of their layoffs, all 26 employees remain with Meta, with separations scheduled to begin on July 22.
The alleged discriminatory practices in the lawsuit are especially focused on employees who were on parental leave. Many had taken time off for pregnancy or childcare, and as such, their work output was significantly reduced during their absences. One plaintiff claimed to have faced discouragement from taking medical leave due to a serious health condition, with a manager warning that doing so could jeopardize his standing during the layoffs. The lawsuit asserts that Meta did not provide any accommodations for this employee's disability.
In its defense, Meta has stated that the claims in the lawsuit "lack merit and are not based on facts," asserting that workforce management decisions were made by individuals rather than automating systems. Interestingly, nearly half of the plaintiffs cited caregiving or pregnancy-related leave as their reasons for absence. Among them, eight women had taken maternity leave, four men had taken parental leave, and one woman had taken leave to care for a family member and subsequently for bereavement.
The lawsuit challenges these layoffs as violations of multiple state and federal laws, which include the Family and Medical Leave Act, the Americans with Disabilities Act, and the Pregnancy Discrimination Act. It also introduces the concept of "disparate impact liability," a principle rooted in civil rights legislation. This principle maintains that employment practices, even if they appear neutral, can still be discriminatory if they disproportionately affect a protected group and are not essential for the job.
Despite the previous administration's attempts to minimize the enforcement of disparate impact liability, the lawsuit against Meta illustrates that such litigation remains a risk for companies in the age of artificial intelligence. Workers retain the right to pursue legal action even if the Equal Employment Opportunity Commission (EEOC) does not take their complaints. Several state laws also prohibit disparate impact discrimination.
The plaintiffs’ legal team argued that Meta's algorithmically assisted selection process resulted in lower performance ratings for employees taking leave, thus disproportionately affecting women, who are more likely to take pregnancy and caregiving leave. The lawsuit references Title VII of the Civil Rights Act, which prohibits employment practices that demonstrate a disparate impact based on gender and other protected characteristics.
As a critical request in the lawsuit, the plaintiffs' lawyers seek to maintain the status quo, preventing the layoffs until arbitration concludes. They emphasize the irreversible harms that employees would face upon separation, including the loss of employer-subsidized health coverage during pregnancy and recovery, the expiration of leave rights, and the forfeiting of unvested equity and immigration benefits.
