Meta is reportedly planning its largest-ever round of job cuts, with potential layoffs affecting 20% or more of its nearly 79,000-strong workforce — translating to over 15,800 positions. According to a Reuters report, the cuts are driven by two converging pressures: the escalating cost of Meta’s AI infrastructure investments, and a deliberate shift toward a leaner, AI-assisted operational model. No final date or exact number has been confirmed, but senior executives have already begun signalling the plans to leadership.
The scale of these reported cuts would surpass every previous round of layoffs in the company’s history, including the twin reductions of 2022 and 2023 that collectively eliminated around 21,000 jobs and were framed under Meta’s “year of efficiency” initiative. This time, the driver is not just operational trimming — it is the direct financial pressure of a company spending at a historically unprecedented rate to dominate the AI race.
Meta Layoffs 2026: What’s Behind the Cuts and What Comes Next
The Financial Weight of Meta’s AI Bets
Meta has committed $600 billion toward building data centres by 2028. Its capital expenditure for 2026 alone is projected to reach as high as $135 billion — nearly double the $72 billion it spent in 2025. On top of infrastructure, Meta has been offering compensation packages reportedly worth hundreds of millions of dollars over four years to attract top AI researchers for its superintelligence team.
The company has also gone on a significant acquisition spree. It acquired Moltbook, a social networking platform built for AI agents, and is spending at least $2 billion to acquire Chinese AI startup Manus. In June last year, it invested $14.3 billion in Scale AI, bringing on its founder Alexandr Wang as chief AI officer.
Zuckerberg has publicly acknowledged the direction the company is heading — noting in January that he was already observing projects that formerly required large teams being handled by a single person with advanced AI tools.
How This Compares to Meta’s Previous Layoffs
| Layoff Round | Year | Jobs Cut | % of Workforce |
|---|---|---|---|
| First “Year of Efficiency” cut | November 2022 | ~11,000 | ~13% |
| Second “Year of Efficiency” cut | March 2023 | ~10,000 | ~12% |
| Reality Labs cut | January 2026 | ~1,500 | ~2% |
| Reported planned cuts | 2026 (pending) | 15,800+ | 20%+ |
The 20% figure, if confirmed, would make this by far the most severe workforce reduction in Meta’s history. Meta spokesperson Andy Stone described the Reuters report as “speculative reporting about theoretical approaches,” stopping short of a direct denial — suggesting internal discussions have taken place but a final decision has not been reached.
The Broader Big Tech Layoff Pattern
Meta would not be alone in this wave. Amazon confirmed one of its largest-ever rounds of layoffs in January, cutting approximately 16,000 jobs — nearly 10% of its workforce. Fintech firm Block went further, eliminating close to half its staff, with CEO Jack Dorsey explicitly pointing to AI tools as the reason for requiring fewer human employees.
Meta has also moved to restructure internally. It recently created a new AI engineering organisation where teams operate with manager-to-employee ratios of up to 1:50, reflecting a deliberate push toward flatter, more AI-reliant structures.
Meta’s AI Model Problems Add Further Pressure
The layoff news coincides with significant turbulence in Meta’s AI development pipeline. Its new foundational model, internally codenamed Avocado, has reportedly underperformed rivals including Google’s Gemini 3.0 in internal tests for reasoning, coding, and writing. Its launch has been pushed back from March to at least May 2026.
Before Avocado, Meta scrapped the largest version of its Llama 4 model — codenamed Behemoth — after it faced criticism for producing misleading benchmark results. Meta assembled an elite internal unit called TBD Lab, led by Alexandr Wang, specifically to build Avocado. So far, the lab’s only public release has been Vibes, an AI video application. With billions committed, a delayed flagship model, and a potential workforce reduction of historic proportions, the pressure on that team is considerable.
