New Delhi [India]: Artificial intelligence is designed to make workplaces more efficient, but can it also unfairly decide who loses their job?
That question is now at the centre of a major legal battle in the United States, where Meta is facing allegations that AI-assisted employee evaluation tools discriminated against workers on medical leave, maternity leave and employees with disabilities during layoffs.
The lawsuit, filed in a California federal court by 26 current and former Meta employees, accuses the Facebook-parent company of relying on AI-driven productivity assessments that allegedly penalised employees who had taken legally protected leave.
Meta has denied the allegations, maintaining that layoff decisions were made by human managers and not by artificial intelligence.
What are the allegations against Meta?
According to the complaint, Meta used AI-assisted workplace tools to rank employees ahead of mass layoffs. The plaintiffs claim these systems relied heavily on productivity metrics, activity scores and digital workplace monitoring to identify employees for termination.
The lawsuit alleges that workers who took maternity leave, medical leave, disability accommodation or family caregiving leave naturally recorded lower activity levels and productivity scores because they were away from work.
As a result, they were allegedly ranked as weaker performers despite their legally protected absences.
Employees argue that these rankings played a role in determining who was laid off.
Which AI tools are under scrutiny?
The lawsuit mentions several internal Meta systems that were allegedly used in employee evaluations, including:
MetaMate, an AI-powered workplace assistant.
Second Brain, Meta's internal knowledge management platform.
AI-generated productivity scores based on employee activity.
According to the complaint, these systems analysed workplace data such as:
Keystrokes
Screen activity
Emails
Internal messages
Browser history
Other digital workplace interactions
The plaintiffs allege that this information was used to generate performance rankings that influenced layoff decisions.
Employees cite specific cases
The lawsuit includes several examples that employees say demonstrate discrimination.
One scientist claims she was informed of her layoff just two days before giving birth.
Another engineer alleges that his performance rating declined after taking leave to recover from an injury.
A manager on approved medical leave claims he was selected for termination just over two weeks after beginning his leave.
The employees argue that these decisions violated federal and state anti-discrimination laws because protected leave was allegedly treated as poor performance.
What are the employees demanding?
The plaintiffs have asked the court to:
Temporarily halt the layoffs while arbitration proceedings continue.
Order an independent audit of Meta's AI systems.
Examine whether the company's AI-assisted evaluation tools contain discriminatory bias.
The lawsuit also alleges that Meta failed to adequately test its AI systems for bias and kept the employee selection process opaque, leaving workers without a clear explanation of why they were chosen for layoffs.
Meta rejects the allegations
Meta has strongly denied the claims.
A company spokesperson said the allegations have "no factual basis" and insisted that workforce management and organisational restructuring decisions were made by people—not artificial intelligence.
The company maintains that employee performance evaluations, rather than AI systems, determined layoff decisions.
Why does this case matter?
The lawsuit has intensified the global debate over the growing role of artificial intelligence in workplace decisions.
While many companies now use AI to measure productivity, automate performance reviews and assist managers, critics argue that algorithms can unintentionally reproduce bias if they rely heavily on activity-based metrics without considering legally protected absences.
Experts say the case could become a significant test of how employers use AI in human resource decisions and whether companies can be held accountable when AI-assisted systems allegedly produce discriminatory outcomes.
Cyber law expert Pawan Duggal said the case raises a broader legal question about accountability in the age of artificial intelligence.
According to Duggal, if AI systems cause harm, responsibility cannot rest solely with software developers. Instead, accountability should extend across everyone involved in designing, deploying and supervising AI systems, with liability determined by the level of human control exercised over the technology.
He argued that existing legal frameworks are not fully equipped to deal with the challenges posed by AI-driven decision-making and called for updated regulations to govern the use of artificial intelligence in workplaces.
The bigger picture
The Meta lawsuit comes at a time when governments across the world are introducing rules to regulate AI in employment decisions.
The outcome of the case could influence how companies deploy AI tools for hiring, performance evaluation and layoffs, while also shaping future regulations on algorithmic transparency, bias testing and employer accountability.
Comments (0)
Be the first to comment!