A United States federal judge in California has declined to intervene to prevent Meta from laying off 26 employees who contend that artificial intelligence systems were used to disproportionately target them for dismissal due to disabilities or medical absences. The ruling from Oakland-based judge William Orrick on Friday means that Meta can proceed with terminations beginning July 22, even as the workers pursue their legal claims through private arbitration proceedings. This decision marks a significant moment in the emerging landscape of AI accountability within corporate America, particularly as the technology giant accelerates its pivot toward artificial intelligence investment.
The judge determined that the affected workers had not demonstrated "irreparable harm" sufficient to warrant an emergency court order blocking the layoffs. Under US legal doctrine, temporary restraining orders to prevent layoffs are exceptionally difficult to obtain, as courts traditionally view job loss as an injury that can theoretically be remedied through monetary compensation if plaintiffs succeed in arbitration. However, Judge Orrick's written decision suggested openness to reconsidering this position should the workers present additional evidence about the specific mechanisms through which AI systems influenced personnel decisions at Meta.
Meta notified approximately 8,000 employees in May—roughly one-tenth of its global workforce—that they would be dismissed as the technology company intensifies its strategic focus on artificial intelligence capabilities. The 26 workers bringing the lawsuit represent a small fraction of those laid off, but their case addresses novel legal terrain that could have far-reaching implications across the technology sector and beyond. These plaintiffs allege that Meta's artificial intelligence tools systematically disadvantaged workers who took medical leave or missed work due to disability-related needs, a claim that directly invokes protected-status discrimination laws.
The workers' legal team argues that Meta deployed multiple AI systems to identify candidates for termination, including a large language model called "Metamate" designed to function as an employee "second brain" that monitored communications and documents. Additionally, the company allegedly utilised productivity scoring mechanisms that examined keystroke patterns, screen activity, email behaviour, and browsing history. Critically, the plaintiffs contend that Meta failed to pause these monitoring systems during vacation periods and legally protected leave, causing AI adoption scores—themselves input metrics for layoff selection—to artificially decline for workers on legitimate absences.
The case represents what appears to be the first significant legal challenge to major American corporate use of artificial intelligence in conducting workforce reductions. This distinction matters considerably, as it places Meta at the centre of emerging questions about the transparency and fairness of algorithmic decision-making in contexts where employee livelihoods are at stake. For Malaysian and Southeast Asian observers, this development signals how technology regulation and worker protections are evolving in advanced economies, potentially influencing future policy discussions and corporate accountability standards in the region.
During Thursday's court hearing, plaintiff advocates highlighted the downstream consequences of the job losses beyond lost salaries and the immediate financial impact. The workers stand to forfeit employer-subsidized health insurance and valuable stock options, creating particular hardship for those with active medical conditions, ongoing pregnancies, or other healthcare needs requiring continuity of coverage. One attorney representing the employees, Barbara Cowan, emphasised to the judge that certain life events—childbirth, medical treatment, bonding with newborns—cannot be reversed or fully compensated through later financial settlements, even if plaintiffs prevail in arbitration.
Meta's legal representatives responded by contending that dismissed workers retain the ability to obtain health coverage independently and that any financial damages suffered would be recoverable through the arbitration process if the company is ultimately found liable. The company has consistently denied wrongdoing and insisted that human decision-makers, rather than artificial intelligence systems alone, made the final determinations regarding which employees would be terminated. This framing attempts to position AI as an analytical tool supporting human judgment rather than as an autonomous decision-making mechanism, a distinction that will likely become central to the legal dispute.
The workers filed their lawsuit anonymously and comprise engineers, managers, researchers, and product designers across Meta's operations. They were notified of terminations in May, with finalisation scheduled for July 22 for many employees and extending into late July or August for others. Notably, these individuals lost access to Meta's internal systems on May 20 and have performed no work for the company since that date, yet formally remain on payroll during the transition period—a status that has influenced the legal argument about what constitutes sufficient harm to warrant emergency court intervention.
The procedural mechanism through which this dispute will be resolved—private arbitration rather than open court proceedings—itself raises questions about transparency and public accountability. Meta's employment agreements require workers to submit disputes to confidential arbitration rather than pursue class-action lawsuits in public courts. While exceptions for temporary relief requests exist in many arbitration agreements, they are ordinarily invoked in cases involving trade secret theft or employee poaching, not in layoffs of at-will employees. The plaintiffs' legal team successfully argued that the temporary relief exception should apply to their circumstances, securing at least the possibility of expedited court review even if the underlying claims will be arbitrated privately.
Judge Orrick's decision to leave open the possibility of reconsidering his determination based on additional evidence about AI use represents a small concession to the workers' concerns, even as he rejected their immediate request for injunctive relief. The judge's suggestion that he might alter course based on further factual findings indicates that the specific technical evidence—how exactly the AI systems functioned, what inputs they used, how their outputs correlated with protected characteristics—will prove determinative. This emphasis on empirical detail about algorithmic operations reflects the judiciary's evolving grappling with artificial intelligence in consequential decision-making contexts.
For regional observers monitoring technology regulation and worker protections, this case illuminates the regulatory gap that currently exists around AI systems affecting employment. Most Southeast Asian countries lack comprehensive legal frameworks addressing algorithmic discrimination in hiring, promotions, and terminations. The Meta litigation demonstrates how workplace artificial intelligence can embed or amplify discrimination, particularly against workers with disabilities or caregiving responsibilities, and signals that legal challenges to such systems are beginning to materialise. As regional economies increasingly adopt AI-driven human resource management tools, policymakers may face pressure to establish clearer accountability standards and transparency requirements.
The workers' pending motion for a preliminary injunction—a longer-lasting temporary order—remains under consideration, and this proceeding may attract additional scrutiny as evidence emerges about Meta's specific AI systems and their effects. Should subsequent filings provide more detailed technical information demonstrating how artificial intelligence contributed to disparate impact on protected groups, Judge Orrick indicated he might revisit the question of whether the layoffs should be halted pending resolution of the underlying claims. This possibility, however limited, preserves a legal avenue for the workers and maintains ongoing judicial scrutiny of Meta's practices even as the company proceeds with its announced terminations. The case will ultimately test whether United States law can effectively constrain algorithmic discrimination in the employment context, a question with significant implications for technology governance globally.
