What happened
26 Meta employees have sued the company in federal court in Oakland, California, claiming it used artificial intelligence systems to pick people for layoffs and that the picks fell disproportionately on workers on medical, parental or family leave, as reported by Talent Canada. They are among the 8,000 employees, about 10 per cent of the workforce, that Meta said in May it would lay off.
The complaint says the company relied on internal AI systems, keystroke and activity-monitoring data, AI token-usage dashboards and algorithmically assisted performance rankings. The plaintiffs argue those scores cannot be built up by someone on protected leave or whose output is reduced by a disability, and that Meta did not pause the process for an individual review that accounts for leave and accommodation.
Meta says the claims lack merit and that workforce decisions were made by people, not AI. The 26 plaintiffs remain employed, with separations set to begin July 22. They are asking the court only to keep them employed while the matter goes to arbitration. The claims rest on US law, including the Family and Medical Leave Act, the Americans with Disabilities Act and the Pregnancy Discrimination Act, and on the disparate impact doctrine. None of this has been decided.
Why it matters if you are hiring
The case is American, but the logic travels. If a ranking counts output, and a person was legally away for part of the period, the ranking can read their leave as weak performance. In Canada, human rights legislation and employment standards leave protections apply in similar ways, and a tool does not remove the employer's responsibility for the outcome.
This lands on HR leaders, operations heads and any manager who builds a cut list from dashboards. It also affects how you hire. Employers that restructure with opaque scoring risk legal exposure and a damaged name with the candidates they want next. Skilled people talk to each other, and how you treat people on leave shows up in your reputation.
If you use productivity data, ask whether it is adjusted for time away. Ask who can explain a score to the person it rates.
One thing to do this week
Pull the criteria behind any current or planned headcount reduction. For each metric, check whether an employee on protected leave or an approved accommodation could score fairly. Where the answer is no, write down how a person will review those cases individually before any decision is final.
Traverse helps employers get access to the best talent across operations, finance, sales and other functions, including the HR and operations people who run workforce changes carefully.