Meta used AI to lay off disabled workers

Tech giant and genocide enabler Meta is accused of using AI software to target disabled people for layoffs in a new court case.

In May, Meta announced they would be sacking around 8,000 people. But a new lawsuit claims the company used a combination of AI tools which gathered data on employees’ performance. These include productivity and activity monitoring. Crucially, the scores of those who took more medical leave were reduced because they were at work less.

Meta uses AI to target disabled employees

Twenty-six employees filed the complaint anonymously on Monday, 13 July, in Oakland, California. The complaint explained:

Those tools draw on inputs… that, by design, cannot be accumulated by an employee who is on protected medical or family leave, or whose output is reduced by a disability.

Whilst Meta could’ve excluded those who took more leave on medical grounds from this, they chose not to:

Meta did not neutralize those inputs for protected leave; did not exclude protected-leave-takers or accommodation-seekers from the selection cohort; and did not pause the system for the individualized, leave- and accommodation-neutral review that the law requires.

This means that not only did Meta fail to protect those on protected leave from an AI that doesn’t understand, but they used that system to disproportionately target disabled people.

As the complaint explains:

The result was that employees who took protected leaves were disproportionately selected for layoff, based on scoring that not only failed to account for their protected leaves, but in effect penalized the employees for exercising their legal rights to these leaves.

The plaintiffs are understandably accusing Meta of violating federal and state laws that ban discrimination or retaliation against workers who have disabilities, take medical leave, or are pregnant. They also say Meta’s AI systems weren’t tested for bias, which would be in violation of recently adopted California and New York City laws.

Employees spied on

The lawsuit explains that Meta quietly launched the monitoring program, and naturally, employees had no option to consent:

The program was announced through a low-visibility internal post—made by an engineer rather than a senior leader, in a secondary group rather than Meta’s official employee-notice channel—with little notice and no consent or click-through acknowledgment; on at least some teams, employees received no consent or acknowledgment prompt at all, and, at least initially, there was no way to opt out.

As the Guardian reports, the twenty-six plaintiffs include a scientist who was on pregnancy leave and then parental leave. She found out she’d lost her job two days before she gave birth. Others include a manager who was let go 16 days into medical leave and an engineer who was given a lower score because of time off due to a work-related injury.

The complaint states:

Meta did not assemble the termination list through the considered judgment of managers who knew the work. Instead, Meta used a constellation of internal artificial-intelligence systems …. to score, rank, and select employees for inclusion on the list.

The thousands of layoffs are supposed to start happening on 22 July, but the plaintiffs are seeking a preliminary ruling from the court, which would block Meta from sacking them all while the case is active. They are also applying for relief that could include back pay and reinstatement of their jobs.

Featured image via the Canary