Twenty-six current Meta employees have initiated legal action against the technology giant, contending that the company deployed artificial intelligence systems to identify candidates for dismissal in a manner that systematically disadvantaged workers taking protected leave. The lawsuit, lodged on July 13 in federal court in Oakland, California, represents a subset of the 8,000 employees—approximately 10 per cent of Meta's total workforce—who received redundancy notices in May. The allegations underscore growing scrutiny of how major technology firms leverage automated systems in workforce decisions, and raise critical questions about algorithmic bias in employment practices that extend well beyond Silicon Valley.
According to the complaint, Meta's selection methodology relied on multiple data sources and computational tools to rank employees, including keystroke monitoring systems, activity tracking mechanisms, AI token-usage dashboards, and algorithmically-assisted performance evaluation frameworks. The lawsuit contends that these systems were fundamentally incompatible with employees exercising their legal rights to medical and family leave, since workers absent from their normal duties inevitably accumulated lower performance metrics. The technical architecture of these evaluation tools meant they could not distinguish between voluntary disengagement and lawful absence, the plaintiffs argue, creating a systematic bias against anyone exercising protected leave entitlements.
The core legal argument centres on a structural flaw in Meta's approach: the company allegedly failed to pause or recalibrate its algorithmic selection process to account for employees on approved leave or those requiring reasonable accommodations for disabilities. Under employment law, employers must conduct individualised assessments that factor in protected status before making termination decisions. Instead, the lawsuit asserts, Meta applied a standardised algorithmic filter that treated all reduced output equally, regardless of whether the reduction stemmed from voluntary underperformance or legally protected absence. This mechanistic approach, the plaintiffs argue, violates the Family and Medical Leave Act, the Americans with Disabilities Act, the Pregnancy Discrimination Act, and the Pregnant Workers Fairness Act.
The demographic composition of the plaintiffs reveals the disparate impact claimed in the lawsuit. Approximately half the 26 employees took leave for pregnancy or caregiving purposes. Eight women had taken maternity or pregnancy-related leave, four men had taken parental leave, and one woman had taken bereavement and family care leave. One plaintiff disclosed that his manager actively discouraged him from taking approved medical leave by warning that doing so would trigger his selection for redundancy—a claim that, if substantiated, would constitute direct evidence of discriminatory intent rather than merely disparate impact.
Meta has rejected the allegations, stating in a company statement that the claims "lack merit and are not based on facts" and emphasising that "workforce management and organisational decisions were and are made by people, not AI." This response echoes a common defence in employment litigation where companies argue that human judgment, rather than algorithmic processes, ultimately determined layoff selections. However, the plaintiffs' legal team contends that even if final decisions involved human approval, the algorithmic pre-screening created a heavily biased pool, effectively delegating the discriminatory gatekeeping function to machines.
The lawsuit invokes the legal doctrine of disparate impact, which holds that facially neutral employment policies can constitute unlawful discrimination if they disproportionately burden protected classes of workers and lack business necessity. This framework is particularly relevant to algorithm-driven decisions because such systems can perpetuate historical biases embedded in training data or create new forms of discrimination through their technical design. The plaintiffs' lawyers argue that Meta's system, by recording periods of protected leave as reduced performance metrics, created a disparate impact on women, who statistically take pregnancy and caregiving leave at significantly higher rates than men.
The Trump administration's recent moves to deprioritise disparate impact enforcement at the federal level add political complexity to this case. Executive orders have directed federal agencies to avoid pursuing disparate impact claims, arguing that such doctrine undermines meritocratic hiring practices. The Equal Employment Opportunity Commission has subsequently dropped discrimination cases for some workers. However, Meta's situation demonstrates that companies remain legally vulnerable to disparate impact litigation despite federal policy shifts, as individual workers retain the right to pursue private lawsuits in federal and state courts, and numerous state laws explicitly prohibit disparate impact discrimination.
For Southeast Asian readers, this lawsuit carries several implications. Many technology companies with significant operations throughout the region—including Malaysia—increasingly adopt similar algorithmic workforce management tools. If courts find Meta liable for algorithmic discrimination, it could establish precedent that influences how multinational corporations calibrate such systems across their global operations. Malaysia's labour laws, while distinct from American statutes, similarly protect workers' rights to medical leave and parental entitlements, and companies deploying AI-driven human resources systems must ensure compliance with local employment regulations.
The case also highlights the broader tension between technological efficiency and human welfare in the age of artificial intelligence. Algorithmic systems can process vast datasets and identify patterns invisible to human managers, but this capability creates risks when the systems incorporate or amplify existing inequalities. The Meta lawsuit suggests that the future of AI in employment will increasingly involve regulatory scrutiny and legal liability for companies that fail to build in safeguards accounting for protected classes and legal entitlements.
The 26 plaintiffs have requested that the court preserve the status quo, maintaining their employment pending arbitration proceedings. Their lawyers emphasise that once separations become final, the harm becomes irreversible: loss of employer-sponsored health coverage during pregnancy and medical treatment, forfeiture of time-limited leave entitlements, loss of unvested equity compensation, and in some cases, immigration consequences for visa-sponsored workers. This argument underscores why the timing of legal intervention matters critically in employment discrimination cases, and why workers affected by mass layoffs must act quickly to protect their rights.
