Meta Platforms Inc. plans to slash about 8,000 jobs starting May 20. That’s roughly 10% of its 78,865 employees. More cuts loom in the second half of 2026. The moves fund a massive AI push, with capital spending set to hit $115 billion to $135 billion this year—nearly double last year’s outlay.
Executives aren’t driven by revenue woes. Meta posted $201 billion in 2025 sales, up 22%. Q4 net income reached $22.8 billion. Free cash flow topped $43.6 billion. Instead, the company eyes AI infrastructure: data centers, GPUs, and tools for Llama models and recommendation engines. A $27 billion joint venture with Nebius bolsters that effort, as detailed in The Next Web.
And the restructuring runs deep. Teams shift into AI “pods” under Chief AI Officer Alexandr Wang’s Superintelligence Labs. Traditional roles vanish. New titles emerge: “AI builder,” “AI pod lead,” “AI org lead.” Maher Saba, head of Applied AI Engineering, wrote in an April 14 memo: “drive a step change in engineering productivity and product quality” and “fundamentally rewiring how we operate.” Engineers move to divisions focused on interfaces, task execution, and data generation.
Reality Labs bears heavy hits. January saw 1,000 to 1,500 cuts there, plus a 30% budget slash. VR game studios shut down. March eliminated 700 across five divisions, including Facebook social, recruiting, sales, and global operations. Specifics include 124 Burlingame positions effective May 22 and 74 in Sunnyvale by May 29. Since 2022, Meta axed about 25,000 roles: 11,000 in November 2022, 10,000 in early 2023, 3,600 in January 2025.
AI Ambitions Clash with Human Costs
Mark Zuckerberg frames 2026 as a year where “the AI wave accelerates even further on several fronts.” CFO Susan Li warns of “significant acceleration in infrastructure expense growth.” Yet a Meta spokesperson dismissed 20% cut reports as “speculative reporting about theoretical approaches,” per Reuters.
Performance reviews tighten. Employees tier into top 20%, middle 70%, lower 7%, bottom 3%. Managers must flag 15-20% as “below expectations.” Top performers snag up to 300% bonuses. Yann LeCun’s late-2025 exit as Chief AI Scientist triggered 600 FAIR researcher cuts. Employees vent on Blind: workplace feels “toxic,” a “crisis of trust.”
Bank of America pegs annual savings at $7 billion to $8 billion. But the tech sector bleeds too. Over 95,000 jobs gone in 2026 across 247 events. Amazon cut 16,000 in January. Oracle eyes up to 30,000, as noted in The Next Web. Meta’s March 700 layoffs hit Reality Labs, recruiting, and sales, amid a new executive stock program, according to The New York Times.
California filings confirm more pain: nearly 200 jobs in Burlingame and Sunnyvale by late May, per New York Post. Reality Labs reorganized into AI pods after 168 Washington state cuts on March 31.
Internal leaks paint a grim picture. X posts from @LayoffAI highlight mandates: 65% of engineers to write 75%+ code with AI by mid-2026. Performance now grades “AI-driven impact.” A gamified app doles badges for quotas. “You don’t build that infrastructure to keep 75,000 engineers employed,” the post warns.
But Meta hires selectively too. AI roles persist amid the purge. Zuckerberg pushes smaller teams powered by AI tools. The company acquired 49% of Scale AI for $14.3 billion, installing Wang as leader.
Broader Tech Reckoning Accelerates
Q1 2026 earnings on April 29 offer a snapshot. Investors watch capex versus returns. Energy bottlenecks loom for AI buildouts, with hyperscalers like Meta racing for power deals.
So what’s next? Layoffs free cash for servers over salaries. Productivity gains promised. Yet morale frays. History repeats: pandemic over-hiring corrected before. Now AI dictates.
Tech insiders see a pattern. CTOs at summits target 70-74% engineering reductions by year-end, per X threads from @TechLayoffLover. “Hiring humans for code is like hiring horses for transportation.” Boards demand it, profitable or not.
Meta’s path tests that bet. Savings fund the future. Humans pay the price. Watch May 20.


WebProNews is an iEntry Publication