Number of internal AI systems used to rate and rank workers on a layoff list, as stated in a lawsuit.
BY Mahnoor | 15-07-2026
Twenty-six workers from Meta Platforms have started a new court case. They say the company used AI software that unfairly picked people with disabilities or those who took medical leave to be fired in big layoffs.
The case was filed in a federal court in Oakland, California on Monday evening. It claims that when Meta cut thousands of jobs earlier this year, it used things like work speed and AI token use as reasons, which hurt people who missed work due to health issues or caring for family.
The workers, who were told in May that their jobs would end starting July 22, are asking a court to stop Meta from firing them while they take their case to a private judge. They say Meta’s rules force each worker to settle disputes alone, but not for urgent help.
A Meta spokesperson said on Tuesday that the workers’ claims are not valid.
“Workforce management and company decisions were and are made by people, not AI,” the spokesperson said.
This lawsuit seems to be the first against a big US company to argue that AI was used to carry out layoffs.
Meta fired 10% of its workers worldwide in May, about 8,000 people, and was planning more job cuts later this year, Reuters reported. Boss Mark Zuckerberg then said he doesn’t expect more company-wide layoffs this year.
These changes are part of a big shake-up as the company puts more money into AI and focuses AI helpers in both its products and how it works inside.
26 people who sued Meta without giving their names say the company broke laws that protect workers with disabilities, those on medical leave, or pregnant women. They also say Meta didn’t check its AI for bias, which goes against new laws in California and New York City.
The people who are suing come from six states and Washington, D.C.
The lawsuit says Meta used AI tools like Metamate, a ‘second brain’ that tracked worker chats and files, and a productivity score from checking keystrokes, screen stuff, emails, and browser history to rate and rank employees for a firing list.
Also watch this :
Leave a comment