A US judge refused to halt Meta's planned layoffs of 26 employees over AI bias claims.
The employees allege AI tools penalised workers who took medical or family leave.
Meta says human managers, not AI, made the final layoff decisions.
Meta has secured a preliminary win in a high-profile legal battle over its use of artificial intelligence in employment decisions, with a US federal judge refusing to block the company from moving ahead with layoffs affecting 26 employees who allege they were wrongly picked for job cuts through AI-based tools.
The decision clears the way for Meta to carry out the terminations, which are expected to start next week, while the broader legal fight proceeds separately through private arbitration. The case has gained attention for the questions it raises about whether AI tools used in workplaces could end up penalising employees who take legally protected leave for medical or family reasons.
4 July 2026
Get the latest issue of Outlook Business
What The Employees Allege
The employees, who have not been named in court documents, claim Meta used several AI-linked systems to help decide which positions would be cut during its recent workforce reduction. They allege these systems worked against staff who had taken time off for medical treatment, disability or caregiving responsibilities.
US District Judge William Orrick, presiding over the case in Oakland, California, turned down the employees' request for an emergency order to stop the layoffs from proceeding while their claims are being heard. He did, however, indicate that the court could revisit the matter if further evidence emerges on "whether and how AI was used" in the layoff process.
The judge's ruling did not address whether the employees' claims hold up. Rather, it turned on the legal standard for emergency relief, with Orrick finding that the plaintiffs had not shown the kind of irreversible harm needed to justify a temporary restraining order.
He added that the court "may reconsider" its findings if the parties present more information about the company's use of AI during the redundancy exercise.
Responding to the order, the employees' legal team said the case was still very much alive. In a joint statement quoted by Reuters, they said, "The Court expressly stated that it may reconsider its determinations based on any additional evidence the parties provide regarding 'whether and how AI was used' in the reduction in force."
Meta's Position
Meta has not issued a public comment on the ruling. The company continues to reject any wrongdoing and says the layoff decisions were made by human managers rather than automated systems.
According to court filings, the employees named in the case are still formally on Meta's payroll but have had no access to internal systems since May and have not been carrying out their regular work. Most of the layoffs are expected to take effect on July 22, with more departures planned through the rest of July and into August.
Inside The AI Tools
The lawsuit identifies several AI-assisted tools it says Meta used while assessing employees for the layoffs. These include Metamate, an internal large language model assistant, a knowledge-sharing tool employees call a "second brain," and a scoring system that reportedly tracked keystrokes, screen activity, emails and browsing history to measure productivity.
The employees also allege they were evaluated using metrics tied to AI adoption and AI token usage, which they say put workers on extended medical or family leave at a disadvantage since they had less opportunity to use the company's AI tools during that time.
The lawsuit further claims that data on employee performance kept being gathered even while staff were on legally protected leave, and that this resulted in lower scores that later played into layoff decisions.
Notably, during arguments this week, attorney Barbara Cowan, representing the employees, said the impact of losing a job at such a moment could not simply be undone. "There's no do-over for bonding with a new baby or giving birth or having active medical treatment," she told the court.
Erin Connell, representing Meta, argued that the employees were losing employer-subsidised insurance rather than access to healthcare itself, and said any financial losses could still be remedied later should the employees succeed in arbitration.
The lawsuit is believed to be the first of its kind against a major US technology company to directly contest the use of AI in employee layoff decisions, a question likely to gain wider relevance as companies increasingly use such tools in workforce management.
