A Reuters report that has been circulating since Friday describes Meta Platforms as internally evaluating workforce reductions that could affect 20% or more of its roughly 79,000 employees, amounting to approximately 15,000 to 16,000 job cuts. The company has not confirmed any specific plans, the timing has not been set, and a company spokesperson, Andy Stone, called the report “speculative reporting about theoretical approaches.”
That language is worth noting, because it neither denies that discussions are happening nor confirms that they have concluded.
The number itself, if it materialises, would make this Meta’s largest single workforce restructuring since November 2022, when CEO Mark Zuckerberg announced 11,000 redundancies, followed by another 10,000 the following spring in what he called the “year of efficiency.”
Those cuts reduced the workforce meaningfully and dramatically improved Meta’s financial performance over the following two years. The company is now applying a similar logic to a new constraint, which is that its AI ambitions cost more than its current cost structure can absorb without adjustment.
The capital expenditure figures provide the context that makes the reported workforce discussions feel less like a rumour and more like a probable outcome. Meta announced in January that its capex for 2026 would be between $115 billion and $135 billion, roughly double the $72 billion spent in 2025, with the increase driven primarily by investments in Meta Superintelligence Labs, large-scale data centre infrastructure and specialised AI talent acquisition. The company has been paying multimillion-dollar packages to attract leading AI researchers, according to various reports, and it has made a series of AI acquisitions, including Manus and Moltbook, across recent months.
Total 2026 expenses are forecast at $162 billion to $169 billion, the scale of which generates the pressure that workforce reductions are designed to relieve. The company still funds these ambitions from advertising revenue, which remains enormous but finite, and the mismatch between revenue trajectory and spending trajectory is what creates the efficiency logic. As AI automates more internal functions, the case for maintaining a large human headcount in areas where AI has reduced the work requirement becomes progressively harder to sustain.
Zuckerberg has described 2026 as “a major year for AI” and has consistently framed the company’s investment as a mission to build “personal super intelligence,” a goal that requires infrastructure spending that would strain almost any balance sheet in the world. The decision to pursue that goal while simultaneously cutting staff in other areas reflects an allocation choice rather than financial distress, a distinction Wall Street has been clear about. Meta’s stock climbed almost 3% on Monday when Reuters’ report first surfaced, which is the market’s way of saying it read the news as evidence of financial discipline rather than corporate distress.
The broader pattern across Big Tech is important for context. Amazon confirmed 16,000 corporate redundancies in January 2026, its largest in the company’s history, with CEO Andy Jassy having previewed the logic months earlier. “As we roll out more Generative AI and agents, it should change the way our work is done,” Jassy wrote to employees in June.
“We will need fewer people doing some of the jobs that are being done today, and more people doing other types of jobs.” Block cut 4,000 people in February. Atlassian announced 1,600 redundancies last week, representing 10% of its workforce. Oracle is reportedly planning between 20,000 and 30,000 job cuts.
A debate has emerged around whether these cuts represent genuine AI-driven efficiency gains or whether, as OpenAI’s Sam Altman and others have suggested, they are partly “AI-washing” where executives use AI transformation narratives to justify headcount reductions that would have happened anyway for financial reasons. Both things can be true simultaneously, and for the employees affected the distinction does not matter much in practice.
Anthropic published research in March suggesting AI currently handles or assists with 75% of daily tasks for computer programmers and is making inroads across customer service and data entry roles. “Workers in the most exposed professions are more likely to be older, female, more educated, and higher-paid,” the paper noted, which paints a more nuanced picture of displacement than the simple narrative of low-skill job losses that dominated earlier debates about automation.
The hyperscalers collectively, including Meta, Alphabet, Amazon, Microsoft and Oracle, have committed approximately $969 billion in data centre investments so far, a figure Moody’s estimates could reach $2 trillion over four years. That commitment is real, it is not reversible at short notice, and it creates a structural case for workforce optimisation that will play out across the sector for several years regardless of what any individual company says about its near-term plans.