Meta to Lay Off Hundreds of Workers Training Its AI

Hundreds of workers in Ireland responsible for refining Meta’s AI models have been informed that their jobs are at risk as the company initiates another major wave of layoffs, based on documents obtained by WIRED. These employees are employed by Covalen, a Dublin-based firm that provides content moderation and labeling services for Meta.

The layoffs were announced during a brief video meeting on Monday afternoon, where workers were not permitted to ask questions, according to Nick Bennett, an employee on the call. “We had a pretty bad feeling [before the meeting],” he says. “This has happened before.” In total, more than 700 Covalen employees could lose their jobs, according to an email reviewed by WIRED. Around 500 of them are data annotators whose job involves checking content generated by Meta’s AI against company policies prohibiting dangerous or illegal material. “It’s essentially training the AI to take over our jobs,” claims another Covalen employee, who spoke anonymously for fear of retaliation. “We take actions as the perfect decision for the AI to emulate.”

The work can involve crafting elaborate prompts to test whether AI guardrails effectively block harmful outputs, such as child sexual abuse material or suicide descriptions. “It’s quite a grueling job,” says Bennett. “You spend your whole day pretending to be a pedophile.”

Last week, Meta announced plans to cut one in ten jobs as part of broader layoffs aimed at increasing efficiency. An internal memo reportedly indicated the cuts were driven by a need to redirect spending toward other areas. While the memo did not specifically mention AI, Meta has recently said it will nearly double its investment in the technology. In January, CEO Mark Zuckerberg stated, “I think that 2026 is going to be the year that AI starts to dramatically change the way that we work.” Covalen employees were told only that the layoffs stemmed from “reduced demand and operational requirements.”

Meta spokesperson Erica Sackin provided a statement: “As we shared in March, over the next few years, Meta will be deploying more advanced AI systems to transform our approach to content enforcement and operations across our platforms, so that it delivers the safety and protection people expect. As we do that, we’ll be reducing our reliance on third-party vendors and strengthening our internal systems.”

This is the second time Covalen has cut staff in recent months. In November, the company announced job cuts reportedly affecting around 400 workers, which led to a strike. Between the two rounds, Covalen’s Dublin headcount is on track to shrink by nearly half, according to the Communications Workers’ Union (CWU), which represents some Covalen employees.

For those affected, finding new work is complicated by a six-month “cooldown period” that bars them from applying to other Meta vendors, says the CWU. “It’s undignified, you know,” says the anonymous Covalen employee. “It’s rude.” Covalen did not immediately respond to a request for comment.

Unions representing the workers are pushing for negotiations over severance terms and hope to meet with the Irish government to discuss AI’s impact on the workforce. “Tech companies are treating the workers whose labor and data helped build AI as disposable,” says Christy Hoffman, general secretary of UNI Global Union. “To fight back, it’s absolutely critical that workers organize and demand notice about the introduction of AI, training linked to employment, and a plan for their futures. Workers should also have the right to refuse to train their AI replacements.”

Some workers caught in the layoffs are skeptical about finding stable employment in a labor market being reshaped by AI and the well-funded companies driving its development. “It’s a universal battle between downtrodden white-collar workers and big capital, really,” claims Bennett. “That normally only goes one way.”