On April 24, Meta announced it would cut 10% of its workforce , roughly 8,000 people , while simultaneously closing 6,000 open roles, and Microsoft revealed the first voluntary buyout program in its 51-year history, together eliminating over 20,000 positions at two companies planning to spend a combined $280 billion on AI infrastructure this year.
The timing was not accidental. Meta's chief people officer Janel Gale sent the internal memo on Thursday morning describing the cuts as helping the company "balance out the other investments we're pursuing." The investments being balanced against 8,000 jobs are enormous: Meta has committed between $115 billion and $135 billion in AI capital expenditure for 2026, nearly double what it spent the previous year. On the same day, Microsoft , which projects $110 billion to $120 billion in AI infrastructure spending for its current fiscal year , disclosed to eligible employees that voluntary retirement was available, a first in the company's history. Both announcements arrived at companies whose combined market capitalization exceeds $4 trillion. Both cited efficiency. Neither mentioned the word AI in the context of the cuts.
Meta's 8,000 layoffs, effective May 20, come on top of the 6,000 vacant roles the company is simply not filling , a combined reduction of roughly 14,000 headcount slots. Microsoft's buyout package targets approximately 7% of U.S. employees, specifically senior directors and below with combined age and tenure totaling at least 70 years, a structure that effectively encourages higher-paid, longer-tenured employees to leave voluntarily. The financial logic is straightforward: a senior director with two decades of tenure costs substantially more than an entry-level AI engineer hired this year, and the transition from the former to the latter is the restructuring both companies are executing through different mechanisms.
Zooming out, the two announcements fit a pattern that has been building since January. According to data compiled by Layoffs.fyi and analyzed by Tom's Hardware, 78,557 tech workers were laid off between January 1 and early April 2026, across 95 companies. Of those, 47.9% , approximately 37,638 positions , were attributed to reduced need for human workers due to AI and workflow automation. AI-attributed layoffs are projected to increase ninefold in 2026 compared to the 55,000 AI-linked cuts reported across all of 2025. Meta, Amazon, Snap, Oracle, Atlassian, and now Microsoft have all made significant cuts this year. The sector has eliminated more than 73,000 jobs in the first quarter alone, a 40% jump from the same period last year.
\h2>Whether AI Is the Cause or the Cover
Cognizant Chief AI Officer Babak Hodjat offered the most honest framing of what is actually happening: "Sometimes AI becomes the scapegoat from a financial perspective, like when a company hired too many, or they want to resize, and it gets blamed on AI." That observation is well-supported by the data. The tech sector massively over-hired during the pandemic , Meta alone grew from 44,000 employees in 2019 to 87,000 by late 2022 , and has been correcting that surplus ever since. The AI narrative provides a convenient and not entirely false explanation for cuts that would have happened anyway. At the same time, Hodjat acknowledged that AI-driven layoffs are coming, and that meaningful productivity gains from modern AI tools will hit the labor market within six to twelve months. The pandemic hangover and the AI restructuring are happening simultaneously, which makes the causal attribution genuinely difficult.
Mustafa Suleyman, Microsoft's head of AI, stated in February 2026 that AI could potentially replace most white-collar jobs within the next 12 to 18 months. That timeline may be aggressive , Cognizant's view is that real productivity gains are still six to twelve months from materializing at scale , but the direction of travel is consistent across every executive making decisions about headcount. LinkedIn data from early 2026 shows AI-related job postings up 340% since 2024 while traditional roles declined 15%. Atlassian laid off 1,600 people while simultaneously announcing 800 new hires in AI-adjacent roles. The pattern is not elimination of work. It is concentration of labor investment at the frontier of the technology, and elimination of roles where AI tools have made the human contribution redundant or marginal.
The Investment Paradox That Is Actually a Strategy
The apparent paradox , spending more on AI while cutting the humans who built the previous generation of products , resolves quickly once you understand what the money is actually buying. Meta's $115 billion in AI capex this year goes overwhelmingly to data center infrastructure, GPU clusters, and energy procurement. The highest-compensated new hires are AI researchers and infrastructure engineers, a population that numbers in the hundreds, not the thousands. The 8,000 jobs being cut are in middle management, operations, recruiting (which shrinks when headcount shrinks), and teams running products that AI is beginning to replace or absorb. The company is not reducing its capacity to build products. It is replacing the labor structure of product development with one that requires fewer humans per unit of output.
For workers in the affected categories, the distinction between "AI as scapegoat" and "AI as genuine cause" matters enormously at the individual level and very little at the structural one. Whether a particular role was eliminated because the company overhired in 2021 or because a language model can now perform 80% of the function, the employment outcome is the same. The roles being created to replace them require different skills, different credentials, and in many cases different people entirely. Anthony Tuggle's assessment, cited across the original reporting, that this represents "a fundamental structural shift rather than a temporary market correction," is the right frame. The tech industry is not going through a downturn. It is going through a reorganization, and the reorganization is designed around a future where significantly fewer humans do significantly more work.
Also read: Chinese AI Tech Turned America's Best AI Against Itself and the Results Are Unsettling • Four Singapore polytechnic students built a $2,500 device that lets paralysed patients speak through blinks • Midjourney's V8 Alpha signals the platform is quietly entering a new era of image and video generation
