IA · 15 July 2026 · 3 min read
When Algorithms Decide Who to Lay Off: Former Employees Sue Meta
In brief: A group of 26 former Meta employees is suing the company, alleging that a complex suite of internal AI tools was used to select individuals for mass layoffs. The lawsuit claims the automated process unfairly targeted workers on protected medical or parental leave.
by Team Mocchi's
The Case: The First Lawsuit Against Algorithmic Layoffs
In Menlo Park, human resource management may have crossed a red line that separates technological efficiency from legal compliance. A group of 26 former employees has filed a lawsuit against Meta, accusing the social media giant of deploying artificial intelligence systems to select workers for termination during its latest restructuring wave. The layoffs, which took place in May, impacted approximately 8,000 employees—around 10 percent of the company's workforce. As first reported by The Verge, this represents the first major class-action lawsuit in the United States to directly challenge the use of predictive algorithms and generative AI tools in executing mass layoffs.
The plaintiffs argue that these automated systems systematically and discriminatorily targeted employees who were on protected medical or parental leave, or who had requested reasonable accommodations for disabilities.
Metamate, Tokens, and "Second Brains": The System Under Scrutiny
According to the complaint filed in the US District Court for the Northern District of California and detailed by Ars Technica, Meta bypassed the qualitative evaluation of direct managers. Instead, the company allegedly relied on what is described as a "constellation" of internal monitoring and AI tools to rank and select workers.
This ecosystem included Metamate—Meta's proprietary internal AI assistant—alongside employee-trained "second-brain" AI agents, keystroke and activity-monitoring software, and dashboards that tracked individual AI token usage. The lawsuit claims that employees were categorized by their adoption rate of these AI tools, with labels such as "AI Native," "AI First," and "AI Enabled."
The structural paradox is that these metrics—especially token consumption, active monitoring logs, and algorithmic performance scores—are impossible to accumulate while an employee is away on legally protected medical or family leave. Consequently, the automated systems interpreted the absence as poor productivity or lack of tech adoption, disproportionately placing these protected individuals on the layoff block.
Meta's Defense and the Legal Challenge of Protected Leave
Meta's response was swift and definitive. Through an official statement, spokesperson Tracy Clayton asserted that the claims lack merit and are not based on facts, reinforcing that workforce management and organizational decisions were and continue to be made by humans, not artificial intelligence.
Nevertheless, the lawsuit exposes a deep-seated gap in automated HR pipelines. The plaintiffs' attorneys argue that Meta's management took no steps to neutralize or adjust the algorithmic scores for workers on approved leaves before finalizing the termination lists. This failure allegedly violated state and federal labor laws, effectively penalizing workers for exercising their legal right to take protected leave.
Mocchi's Take
The Meta lawsuit serves as a critical warning for any business looking to automate its human resources processes. In the European and Italian context, this scenario is bound by even stricter guardrails: the EU AI Act explicitly classifies AI systems used in HR—ranging from recruitment and performance evaluation to termination decisions—as "high-risk" applications. Relying on quantitative metrics like AI token usage or activity tracking without a robust, well-documented "human-in-the-loop" framework exposes organizations to massive regulatory penalties and irreversible reputational damage. At Mocchi's, we believe that while AI is an extraordinary catalyst for workflow optimization, the assessment of human talent and career-altering decisions must remain firmly anchored in human empathy, ethical governance, and full compliance with labor protections.