A U.S. judge recently dismissed a lawsuit aimed at halting Meta’s layoffs of 26 employees. The case claims that the company’s AI tools unfairly targeted these workers due to their disabilities or medical leave.
District Judge William Orrick, based in Oakland, California, stated in his ruling that Meta’s layoffs would proceed starting on July 22, even as the legal claims will be examined through private arbitration.
The judge noted that the workers did not demonstrate that losing their jobs constituted the “irreparable harm” necessary to issue an immediate order against the layoffs.
Meta has consistently denied any wrongdoing, asserting that layoff decisions were made by human personnel.
In May, the company informed around 8,000 of its employees—about 10 percent of its total workforce—that they would face job losses as part of a significant investment in AI technology.
The lawsuit, filed earlier this week, contends that Meta used AI tools measuring productivity and usage metrics to determine which positions to eliminate, effectively penalizing those who took medical leave or cared for family members. Plaintiffs argue that performance evaluations also factored in these AI metrics.
This case appears to be the first legal challenge against a major U.S. corporation over the alleged use of AI in layoffs.
“You can’t start over.”
The plaintiffs sought a temporary restraining order to stop Meta from continuing its layoffs while their case is underway in arbitration. A longer-term injunction is pending, with Judge Orrick indicating he might decide on it next month.
During the hearing, lawyers for the plaintiffs emphasized the potential loss of jobs, salaries, stock options, and health insurance, raising concerns about jeopardizing medical care for pregnancy and other conditions.
One attorney, Barbara Cowan, highlighted the significant emotional impact of losing access to critical time off for family bonding, stating, “There is nothing undone about bonding with a newborn baby.”
In response, Meta’s Erin Connell clarified that while workers would lose employer-subsidized insurance, they would not lose their entire health coverage. She noted that these damages could be claimed later if the plaintiffs are successful in arbitration.
Workers indicated that Meta’s arbitration agreement requires individual resolution of disputes, but they argue this doesn’t prevent requests for temporary relief.
Generally, workers at large firms must sign such agreements, which often push them to handle claims separately rather than in collective lawsuits. Advocates for employees argue that this system tends to favor the employer.
While exceptions for seeking temporary relief from arbitration are recognized, they typically apply to serious matters like trade secret theft—not to layoffs.
The plaintiffs, who are anonymous, include various professionals—engineers, executives, researchers, and designers. The layoffs that began in May are set to conclude officially on July 22 for many, with some going through in late July or August, according to legal documentation.
Though they remain on the payroll, these employees lost system access on May 20 and haven’t worked since. The employees allege that Meta utilized a suite of internal AI tools to assess and rank them for termination, including a language model called “Metamate” and other systems monitoring employee activities.
According to the complaint, these tools failed to pause during employees’ vacations or legally protected leaves, leading to lower AI adoption scores, which were part of the criteria for layoffs.
