Meta's AI picked who to fire. 26 employees say it targeted people on medical leave.

26 Meta employees just sued the company.

Their claim: Meta used "a constellation of internal AI systems" to score, rank, and select who got laid off in its 8,000-person reduction in force.

The AI used performance ratings, productivity metrics, AI-token consumption, and output scores to decide.

Here's the problem.

Employees on medical, parental, or disability leave cannot accumulate those scores by definition.

Meta did not neutralize those inputs.

The result: people exercising their legal right to protected leave were disproportionately selected for termination.

One scientist was chosen while on pre-birth pregnancy leave.

An engineer's rating dropped because of "broken time" from an injury.

A manager was demoted after medical leave, then laid off weeks into his second leave.

Meta's response: "Workforce management decisions were and are made by people, not AI."

But the lawsuit says the AI did the scoring. The AI did the ranking. The AI built the termination list.

Violations alleged: ADA, FMLA, Pregnancy Discrimination Act, Pregnant Workers Fairness Act, Title VII.

Separations begin July 22.

If your organization uses AI-driven performance scoring, workforce analytics, or algorithmic ranking for any HR decision — audit those systems today.

The inputs that look neutral to a machine are often anything but neutral to a lawyer.

SOURCE: https://www.hrdive.com/news/metas-ai-based-layoffs-allegedly-targeted-workers-who-had-taken-protected/825325/

VERIFIED: AP News, HR Dive, CNBC, USA Today, The Guardian

SIGNAL: This is the first major lawsuit testing whether AI-driven layoff selection violates federal employment law. Every enterprise using algorithmic workforce decisions is now on notice.