Former employees of Meta Platforms Inc. have initiated legal action against the technology giant, alleging that the company utilized artificial intelligence algorithms to facilitate discriminatory termination practices. The lawsuit, filed in California, centers on claims that Meta’s internal performance management systems—which incorporate AI-driven metrics—disproportionately targeted specific groups for layoffs during the company’s workforce reduction cycles. This case highlights growing concerns regarding the intersection of automated decision-making and employment law in the technology sector.
According to court filings, the plaintiffs argue that the reliance on AI for performance evaluations and subsequent redundancy decisions lacks the transparency required to ensure non-discriminatory outcomes. The legal challenge seeks to hold the company accountable for how its proprietary tools influence human resources decisions, particularly during high-volume restructurings. As of the latest court dockets, the litigation is in its preliminary stages, with the plaintiffs seeking class-action status to represent other workers who believe they were unfairly impacted by these automated systems.
Legal Scrutiny of Automated Employment Decisions
The core of the complaint revolves around the use of algorithmic management, a practice that has become increasingly common among Silicon Valley firms aiming to quantify employee productivity. The plaintiffs allege that Meta’s “Performance Rating” systems, which were adjusted during the company’s “year of efficiency,” relied on data points that may have codified existing biases. Under California’s Fair Employment and Housing Act (FEHA), employers are prohibited from making employment decisions based on protected characteristics, and the lawsuit contends that the AI-driven process serves as a veil for potential violations of these statutes (California Civil Rights Department, Employment Discrimination Guidelines).
Legal analysts following the case note that the burden of proof will likely rest on the plaintiffs’ ability to demonstrate that the algorithmic outcomes resulted in a “disparate impact.” This is a complex legal standard in the United States, requiring evidence that a facially neutral policy—in this case, an AI performance metric—adversely affects a protected class at a significantly higher rate than others. The Equal Employment Opportunity Commission (EEOC) has previously issued guidance warning that the use of software in hiring and firing can violate federal civil rights laws if it excludes individuals based on race, color, religion, sex, or national origin (EEOC Guidance on Automated Systems).
Meta’s Stance and Corporate Policy
Meta has consistently maintained that its performance evaluation processes are designed to be meritocratic and are subject to human oversight. In various public statements regarding its workforce management, the company has emphasized that while AI tools are used to aggregate performance data, final decisions regarding employment status involve management review. However, the plaintiffs’ legal counsel asserts that the “human in the loop” aspect is often a formality, with managers deferring to the AI’s ranking outputs without independent verification.
The company’s workforce reductions, which saw thousands of employees depart in 2022 and 2023, were part of a broader shift in corporate strategy initiated by CEO Mark Zuckerberg. During these periods, the company transitioned toward a flatter organizational structure, which included a more rigorous focus on individual productivity metrics. While Meta has not issued a specific rebuttal to the individual allegations in this lawsuit, it has historically defended its HR practices as compliant with all applicable labor laws and internal equity standards.
The Future of AI Transparency in the Workplace
This lawsuit arrives as regulators worldwide are stepping up efforts to govern the use of artificial intelligence in the workplace. The European Union’s AI Act, for instance, classifies AI systems used in employment, worker management, and access to self-employment as “high-risk,” imposing strict transparency and human-oversight requirements on companies operating in that market (European Commission, AI Act Overview). While the California lawsuit is governed by U.S. law, industry observers suggest that the outcome could influence future legislative efforts to regulate algorithmic management domestically.
For current and former tech employees, the case serves as a test of how much transparency is legally required when AI is used to justify significant professional disruptions. If the court allows the case to proceed as a class action, it could force Meta to disclose sensitive information regarding the architecture of its performance-tracking algorithms. Such a discovery process would be a milestone in the ongoing debate over whether software-driven management is a neutral tool for efficiency or a hidden engine for systemic bias.
The next phase of the litigation will involve pre-trial motions and discovery, where both parties will exchange evidence regarding the specific metrics used by Meta’s internal systems. No dates for a trial have been set, and the court is currently evaluating the plaintiffs’ request for class certification. Interested parties can monitor the progress of the case through the California Superior Court’s public access portal for civil filings.
Maria Petrova, Editor of World, covers the intersection of technology, labor rights, and international policy. For more updates on this case and other developments in global labor law, please share your thoughts in the comments or subscribe to our newsletter.
Worth a look