Mark Zuckerberg has a simple, albeit brutal, math problem. He's spending billions on AI chips, and he's taking that money directly out of the payroll budget. During a recent internal town hall, the Meta CEO didn't mince words. He explained that the company essentially has two massive buckets of spending: people and infrastructure. To fill the infrastructure bucket with the hardware needed for "Superintelligence Labs," he's emptying the people bucket.
This isn't about robots taking jobs because they're smarter. It’s about the raw cost of the electricity and silicon required to build those robots. Meta is cutting roughly 8,000 jobs—about 10% of its workforce—starting May 20, 2026. This comes on top of 6,000 cancelled open roles. If you’re waiting for the "Year of Efficiency" to end, don't hold your breath. Zuckerberg refused to rule out more cuts later this year, telling staff he doesn't have a "crystal ball" for the next three years.
The Tradeoff Between Silicon and Salaries
The scale of Meta's spending is hard to wrap your head around. The company just hiked its 2026 capital expenditure guidance to a staggering range of $125 billion to $145 billion. For context, that’s nearly double what they spent in 2025. Most of this cash is flowing into data centers and Nvidia H100s (and their successors) to power Llama models and new AI agents.
Zuckerberg’s logic is cold. "If we’re investing more in one area... that means we have less capital to allocate to the other," he told employees. He's essentially choosing GPUs over humans. While he claims that internal AI tools making workers more efficient isn't the primary driver of these layoffs, the reality is that the company is being "fundamentally rewired."
Meta is reorganizing into AI-focused "pods." If your job doesn't fit into a pod dedicated to building or supporting AI, your seat is at risk. They’ve even created new titles like "AI Builder" and "AI Pod Lead." It’s a total shift in identity.
Why the Stock Market Loves the Pain
You’d think firing 10% of your staff would be a sign of trouble. Instead, Meta's revenue surged 33% in the first quarter of 2026. Net income jumped 61% to $26.7 billion. The company is making more money than ever, yet it’s still handing out pink slips.
Investors aren't looking at the human cost; they’re looking at the $9 trillion market cap goal Meta set for 2031. Executives have massive stock incentives tied to this target. Chief Product Officer Chris Cox and CTO Andrew Bosworth could see payouts nearing $1 billion if they hit those numbers. Cutting the largest expense—people—while doubling down on the technology of the future is exactly what Wall Street wants to see, even if it leaves the internal culture in tatters.
Surveillance as Training Data
There's a darker side to this restructuring that isn't getting enough attention. To fuel their AI ambitions, Meta started the "Model Capability Initiative." It’s basically corporate spyware. The software tracks every keystroke, mouse movement, and screenshot on US employees' work computers.
The goal? To train AI agents to mimic how humans work. Essentially, the people still at Meta are being forced to train their own automated replacements. Andrew Bosworth told staff there’s no way to opt out. It’s a "train the bot or leave" ultimatum.
This has caused a predictable meltdown on internal forums. Employees are understandably furious that their daily movements are being harvested to build the very tools that make their roles redundant. It’s one thing to lose your job to a market shift; it’s another to spend your final months teaching a machine how to do your specific tasks.
What This Means for Your Career
If Meta is the bellwether for Big Tech, the message is clear: being "good at your job" isn't enough anymore. You have to be "AI-relevant."
Meta is poaching elite AI researchers with packages worth $1.5 billion. They’re firing recruiters, sales staff, and mid-level managers to afford a handful of "superstar" engineers. The middle class of Silicon Valley is being hollowed out to fund a high-stakes arms race.
Practical Steps for Tech Workers
Don't wait for a town hall to realize your department is a "cost center." If you're in tech or a related field, you need to pivot immediately.
- Audit your "AI-adjacent" skills. Can you lead an AI pod? Can you integrate LLMs into your existing workflow? If not, you're a target for the next "rebalancing."
- Watch the Capex. When a company starts talking about "infrastructure investment" and "compute constraints," it’s a coded warning that the payroll budget is under pressure.
- Diversify your expertise. The jobs being saved at Meta are those that directly contribute to "Personal Superintelligence" or core AI infrastructure. Generalist roles are the first to go.
Zuckerberg is betting the entire house on AI. He's proven he's willing to burn through thousands of loyal employees to get there. If you're working at a major tech firm, assume the "Year of Efficiency" is now the permanent state of business. Get your resume ready and start learning how to manage the machines, because the people who just "do the work" are being priced out by the chips.
