Silicon Valley tech giant Meta faces a new legal challenge. According to reports, 26 laid-off employees have filed a lawsuit in federal court in Oakland, California, accusing the company of using AI-driven algorithms to unfairly screen employees in a recent round of layoffs, resulting in employees with disabilities or who had taken medical leave being disproportionately listed as targets for layoffs.
According to the indictment, these employees received layoff notices in May of this year and their positions would be terminated. The plaintiff believes that Meta relied too much on data such as productivity indicators and artificial intelligence usage in its layoff decision-making process, and that these evaluation mechanisms did not take into account the special circumstances in which working hours were affected due to physical health conditions or medical leave. The lawsuit states that Meta used a series of AI tools to build the layoff list, including a "Metamate" tool that monitored internal employee communications and a system that measured employee productivity by actively scanning employee keystrokes, screen content, email and browser history. Plaintiffs argue that their medical conditions and time off experience were unfairly “penalized” in these data-driven metrics, leading directly to layoff decisions.
The core claim of this lawsuit is to ask the court to issue an injunction to prevent further layoffs that will be officially implemented on July 22. The plaintiff cited relevant federal and local laws in the United States, which clearly state that companies are not allowed to discriminate or retaliate against employees who have certain medical conditions, are pregnant, or are on medical leave.
At the same time, Meta’s current AI strategy is in a period of adjustment. Although CEO Mark Zuckerberg recently stated that there will be no further large-scale layoffs this year, Meta still maintains extremely high intensity in investment in AI infrastructure. Currently, the company is actively promoting the "Meta Compute" plan, which aims to rent inference computing power to external parties. In addition, Meta plans to start manufacturing its application-specific integrated circuit (ASIC) code-named "Iris" in September this year. The chip is designed by Meta in cooperation with Broadcom and will be produced by TSMC. It is understood that Meta is currently one of the companies with the highest investment in the AI field. It aims to deploy 7 gigawatts of computing power by the end of 2026, with capital expenditures expected to be as high as $145 billion, and plans to double this capacity to 14 gigawatts next year. At the product level, Meta also recently launched a paid version of its flagship AI model "Muse 1.1" in an attempt to use low-cost reasoning as its core competitive advantage to cope with challenges from competitors such as Anthropic and OpenAI. Currently, this legal lawsuit regarding algorithmic discrimination in layoffs is becoming an important case for the outside world to examine the boundaries between the internal management and artificial intelligence of technology giants.