Mark Zuckerberg’s Meta is quietly preparing for what could be the largest workforce reduction in the company’s history, with internal discussions centering on layoffs that would eliminate 20% or more of its global headcount, according to a Reuters report citing three people familiar with the matter.
For a company that employed nearly 79,000 people as of December 31, 2025, that translates to roughly 16,000 jobs. To put that in perspective: it would surpass the combined 21,000 positions Meta eliminated during its so-called “year of efficiency” in late 2022 and early 2023, the period that was supposed to have already right-sized the company.
Meta spokesperson Andy Stone dismissed the reporting as “speculative reporting about theoretical approaches.” But the Reuters sources paint a different picture: senior executives have already told other leaders across the company to start planning how to pare back their teams. No final timeline or exact headcount has been set, but the directive to start preparing is itself a signal that this is more than theoretical.
The $135 Billion Question
The math behind these potential cuts isn’t complicated. Meta told investors in January that it plans to spend between $115 billion and $135 billion on capital expenditures in 2026, nearly double the $72 billion it spent in 2025 and roughly triple its $39 billion outlay in 2024. That spending is almost entirely directed at AI infrastructure: data centers, servers, computing power, and the talent to build what Zuckerberg calls “personal superintelligence.”
The company has committed to two massive data center projects. “Hyperion,” a $50 billion facility in Louisiana designed for 5 gigawatts of capacity, and “Prometheus,” a training cluster in Ohio expected to come online late this year. Meta has also gone on an AI acquisition spree, spending $14.3 billion to hire Scale AI founder Alexandr Wang as chief AI officer last June, acquiring social networking platform Moltbook, and reportedly spending $2 billion to buy Chinese AI startup Manus.
Wall Street initially cheered the spending. When Meta announced the capex guidance alongside blowout Q4 2025 earnings (revenue of $59.9 billion, up 24% year-over-year, with EPS of $8.88 beating estimates), shares popped 10% after hours. The logic was straightforward: Meta’s advertising machine was throwing off enough cash to fund any AI ambition Zuckerberg could dream up.
But that enthusiasm has cooled. META shares closed Friday at $613.71, down 3.83% on the session and roughly 23% below their 52-week high of $796.25. The stock has dropped more than 8% over the past month alone. Investors are starting to ask the uncomfortable question: what exactly is all that spending producing?
AI Ambitions Meet AI Setbacks
The answer, so far, is mixed at best. Meta abandoned its “Behemoth” AI model last year after it failed to meet internal benchmarks. The company’s current flagship effort, codenamed “Avocado,” is reportedly lagging behind schedule. Reports emerged on Friday that Meta is delaying the model’s launch, adding to the narrative that the company is spending aggressively on AI infrastructure while struggling to produce models that can compete with OpenAI, Google, and Anthropic at the frontier.
That gap between spending and results is what makes the layoff calculus so pointed. Meta isn’t just cutting costs to improve margins. It’s cutting human workers to fund machines that are supposed to replace them, while simultaneously failing to build the AI systems that would justify the replacement. It’s a circular logic that only works if you believe the AI will eventually catch up to the investment.
Zuckerberg himself has leaned into this framing. He has suggested publicly that AI tools are enabling smaller teams to accomplish work that previously required large departments. That talking point has become standard across Silicon Valley, but it’s worth interrogating: if AI is truly making workers redundant, why is Meta simultaneously hiring expensive AI researchers and engineers at premium salaries?
The “AI-Washing” Debate
Meta is far from alone in using AI as the justification for massive layoffs. Amazon has cut roughly 30,000 corporate roles since October, including 16,000 announced in January and potentially 14,000 more in a reported second wave. Fintech company Block slashed its workforce by nearly half. The pattern is consistent: blame AI for eliminating jobs, spend on AI infrastructure, and promise that the efficiency gains will eventually materialize in the earnings.
Even OpenAI CEO Sam Altman has pushed back on this trend, suggesting that many of these cuts amount to “AI-washing,” where executives use artificial intelligence as convenient cover for layoffs driven by over-hiring during the pandemic, slowing growth, or plain old cost-cutting. TechCrunch noted the irony: the CEO of the company most responsible for the AI hype cycle is cautioning others against using AI as an excuse.
For Meta specifically, there’s a credibility question. The company already went through a brutal restructuring cycle. It cut 11,000 jobs in November 2022, roughly 13% of its workforce. Four months later it cut another 10,000. In January 2025, it pushed out 3,600 employees labeled “low performers.” And just two months ago, in January 2026, it eliminated roughly 1,500 positions from its Reality Labs division as part of a strategic retreat from virtual reality toward AI and wearable technology.
Each round of cuts was framed as the last necessary adjustment. Each time, the message was that Meta had found its right-sized footing. A 20% reduction in 2026, on top of everything that came before, suggests the company either didn’t learn from its previous overcorrections or is genuinely pivoting its entire operational model toward a future where AI does most of the work. Either explanation should make investors nervous.
What’s Actually at Stake
The financial picture remains strong on paper. Meta generated over $200 billion in revenue for 2025. Its advertising business, powered by Facebook, Instagram, WhatsApp, and Messenger, continues to grow at rates that would be enviable for a company a fraction of its size. Q4 ad revenue surged 24% year-over-year, driven by 18% higher ad impressions and robust demand.
But the tension between that cash-generating machine and the AI spending spree is becoming harder to ignore. If Meta spends $135 billion on capex this year, that doesn’t leave much room for error. The company funded part of its AI push with $30 billion in bonds and a $29 billion infrastructure financing deal with a consortium including PIMCO and Blue Owl Capital. It’s not operating from pure cash flow anymore. It’s leveraging up to chase a technology bet that hasn’t paid off yet.
Meanwhile, Reality Labs continues to hemorrhage money. The division recorded a $4.4 billion loss in Q3 2025 alone on just $470 million in revenue. Cumulative losses since late 2020 have exceeded $70 billion. The January layoffs and studio closures signal that even Zuckerberg has accepted the metaverse isn’t delivering returns on the timeline investors need.
The Bigger Picture
If these layoffs materialize at the reported scale, they will represent more than a company restructuring. They will be a statement about how Big Tech sees its own future: fewer humans, more machines, and a willingness to bet enormous sums on a technology whose ultimate productivity gains remain largely theoretical.
For the roughly 16,000 workers who could be affected, the “speculative” and “theoretical” framing from Meta’s spokesperson rings hollow. Senior leaders have been told to prepare. The planning is underway. The question isn’t whether Meta will cut jobs. It’s how many, how soon, and whether the AI systems they’re building will actually justify the human cost.
Meta’s next earnings report is expected on April 28. By then, the company may have made its decision. And the market, which has already started pricing in skepticism with an 8% decline over the past month, will be watching to see whether Zuckerberg’s $135 billion AI gamble is a visionary investment or the most expensive efficiency program in corporate history.
