A US judge on Friday rejected a bid by 26 employees of Meta Platforms to block the tech giant from laying them off while they pursue claims that they were targeted for job cuts by the company’s AI-powered tools because they have disabilities or took medical leave.
District Judge William Orrick in Oakland, Calif., in a written order said he would not stop Meta from carrying out the layoffs beginning July 22 while the merits of the workers’ novel legal claims are decided in private arbitration.
The judge said the workers could not show that losing their jobs amounted to the “irreparable harm” required for him to issue an emergency order blocking the layoffs.
Meta and lawyers for the plaintiffs did not immediately respond to requests for comment. The company has denied wrongdoing and said that decisions involving the layoffs were made by humans.
Meta in May notified nearly 8,000 employees, or about 10% of its global workforce, that they were losing their jobs as the company doubles down on its investments in AI.
The lawsuit filed on Monday claims that in selecting jobs to cut, Meta relied on AI tools that measured productivity and AI token usage, disadvantaging people who missed work because of medical conditions or to care for family members. The company also relied on performance reviews based in part on employees’ adoption of AI, the plaintiffs said.
The case appears to be the first against a major US company to challenge the alleged use of AI in conducting layoffs.
‘No do-over’
The plaintiffs had asked Orrick for a temporary restraining order blocking Meta from completing its layoffs while they pursue their claims in private arbitration. Their motion for a preliminary injunction, a longer-lasting temporary order, is pending and Orrick during the hearing on Thursday said he would likely rule on it next month.
Lawyers for the plaintiffs said during a hearing on Thursday that along with their jobs and salaries, the workers stood to lose valuable stock options and their health insurance, imperiling their medical care for pregnancies and other conditions.
“There’s no do-over for bonding with a new baby or giving birth or having active medical treatment,” one of the lawyers, Barbara Cowan, told Orrick.
Erin Connell, who represents Meta, countered that the workers were losing only employer-subsidized insurance, and not their coverage altogether. Those are the typical kinds of damages that can be recouped later on if the plaintiffs win their cases in arbitration, Connell said.
The workers say Meta’s agreements require employees to arbitrate workplace disputes individually, but do not apply to requests for temporary relief.
Most workers at large companies sign arbitration agreements, which generally require employees to pursue workplace claims individually rather than through class actions in court. Companies say arbitration can provide a faster, cheaper alternative to litigation, while critics say it often favors employers and discourages workers from bringing claims.
Exceptions in arbitration agreements for temporary relief are common, but they are typically invoked in cases involving the alleged theft of trade secrets or the solicitation of clients or employees, and not layoffs of at-will employees.
The plaintiffs, who filed the lawsuit anonymously, include engineers, managers, researchers and designers. They were notified in May of the layoffs, which are scheduled to be finalized on July 22 for many workers and later in July or August for others, according to court filings.
Laid-off workers remain on the payroll but lost access to Meta systems on May 20 and have not performed work for the company since, Meta said in court filings.
They claim that Meta used a number of internal AI-assisted systems to score and rank employees on a termination list. Those included a large language model assistant known as “Metamate,” an employee-trained “second brain” that tracked workers’ communications and documents, and a productivity score drawn from scanning keystrokes, screen content, emails and browser history, according to the lawsuit.
Meta did not pause these systems while employees were on vacations and legally protected leave periods, and their AI adoption scores used as inputs for layoff selection dropped as a result, the plaintiffs said.
