It’s only May, and the tech industry has already handed pink slips to more than 92,000 people. Spread across 98 companies, the job cuts have been relentless—and April turned into the worst single month for announced tech-job reductions in at least two years, with 45,800 employees affected. Meta, Snap, Microsoft, Oracle, Block, Amazon, Nike, and GoPro all moved within weeks of each other, creating a kind of layoff pile-on that the industry hasn’t seen since the post-pandemic reckoning of 2022 and 2023.The stated reason, almost universally, is AI. Companies are spending eye-watering amounts on data centers and chips, and they need to find that money somewhere. Workers, it turns out, are the somewhere. The pitch to investors is straightforward: we’re getting lean so we can get smart. Whether it’s actually working—or whether AI is just a convenient excuse to do what overstuffed tech giants were eventually going to do anyway—is a much messier question.
8,000 at Meta, 1,000 at Snap, buyouts at Microsoft: April’s brutal scorecard
Meta moved the hardest. On April 22, the company told its employees that 8,000 of them—roughly 10% of the global workforce—would be let go on May 20. Another 6,000 open roles would simply stay unfilled. The company has pledged $135 billion in capital expenditure for its latest AI push, and someone has to pay for all those data centers. This time, it’s the people.CEO Mark Zuckerberg addressed employees directly at an internal town hall, offering what he described as the honest version of the math. The company has two major cost centers: compute infrastructure and people. Spend more on one, you have less for the other. He didn’t sugarcoat it, and he didn’t promise it was over. Asked about future cuts, he said the company would “be able to share more soon”—which, in corporate speak, rarely means good news.Snap followed a week earlier with a cut that was arguably even more dramatic in proportional terms. The Snapchat parent said it would eliminate around 1,000 jobs—16% of its full-time workforce—and leave another 300 open roles permanently unfilled. CEO Evan Spiegel pointed to AI as the enabler, noting that the technology now generates more than 65% of Snap’s new code, allowing the company to operate with smaller, more focused teams. The company is targeting over $500 million in annualized savings by the second half of the year.Microsoft took a quieter approach—offering voluntary buyouts to roughly 7% of its US employees rather than announcing a hard layoff. The financial incentive is meant to make departures feel like a choice. For employees closer to retirement, it often is. But the headcount reduction is real either way, and Microsoft’s top finance executive made clear that total employee numbers would fall this year as the company pursues what it’s calling “pace and agility.” If not enough employees accept the offer voluntarily, actual layoffs are expected to follow.
30,000 gone at Amazon, AI debt choking Oracle, and Block’s radical 40% cut
Amazon’s cuts have been staggering in raw numbers, and unlike some companies on this list, they didn’t happen all at once. The company announced in January that it was eliminating around 16,000 corporate roles globally—and that was already the second round. Back in October, it had shed another 14,000 roles. That’s 30,000 white-collar positions gone in roughly six months, framed internally as an effort to reduce bureaucracy. Amazon SVP Beth Galetti described it as necessary restructuring, but at that scale, it’s hard to call it anything other than a fundamental rethink of how the company wants to be staffed.Oracle’s story fits squarely into the broader pattern of AI spending that hasn’t yet paid for itself. The company has been pouring money into AI infrastructure—data centers, computing capacity, the whole build-out—and the returns simply haven’t materialized at the pace the spending demands. That gap between what’s going out and what’s coming in has forced the company to find savings elsewhere, and headcount is the most immediate lever available. It’s the same logic driving cuts at Meta and Microsoft, just with less fanfare around it. Oracle even has a related debt problem: Wall Street has been increasingly vocal about the company’s AI-related borrowing, with analysts flagging that its debt load is pushing toward uncomfortable territory. Cutting staff, in that context, isn’t a vision statement—it’s a balance sheet necessity.Block made arguably the most dramatic single announcement of anyone. In February, CEO Jack Dorsey told employees that 40% of the company’s workforce—more than 4,000 people—would be cut. The parent of Square and Cash App had a headcount-to-revenue ratio that had grown increasingly hard to justify, and Dorsey’s approach to the announcement was notably direct: “We’re not making this decision because we’re in trouble,” he wrote. The line got a lot of attention, partly because it was unusually candid, and partly because companies that genuinely aren’t in trouble rarely feel the need to say so.
Big tech is spending $674 billion on AI, and workers are footing the bill
The numbers behind this year’s layoffs have one common thread—the sheer scale of AI infrastructure spending.Alphabet, Meta, Amazon, and Microsoft are collectively expected to spend $674 billion on capital expenditures this year. That figure is more than double what the same group spent just two years ago, when AI spending was already considered aggressive. Amazon is expected to actually burn cash this year. Meta’s capital spending will account for more than half of its annual revenue.These companies are, in effect, in an arms race they’ve decided they can’t opt out of. Each one is betting that spending the most on AI infrastructure now will put them ahead in a race where second place may not be viable. The workforce reductions are partly a way to generate the capital headroom to keep spending—and partly a way to show investors that management is being disciplined, even as the capital expenditure numbers spiral upward.But a growing number of voices are pushing back on the AI explanation. Marc Andreessen, a prominent venture investor and Meta board director, said recently that the wave of big-company layoffs is fundamentally a correction for pandemic-era overhiring—and that AI is simply the story companies are telling to make it sound visionary. “Now they all have the silver bullet excuse: ‘Ah, it’s AI,'” he said. Sam Altman, CEO of OpenAI, made a similar point at a March conference, saying companies are blaming AI for layoffs “whether or not it really is about AI.”
23% at GoPro, three of four engineers at Tailwind: No safety net at the smaller end
Not every company in this story has thousands of employees to absorb the shock.GoPro cut 145 workers in early April—which is 23% of its entire workforce, gone in one move. The action camera maker is facing a brutal combination of pressures: slowing consumer demand, higher memory costs driven by AI chip competition, and tariff-related cost increases that have squeezed margins from both sides.And then there’s Tailwind, the web development tool, which may have produced the starkest quote of the entire year. In January, CEO Adam Wathan posted a GitHub comment explaining that the company had let go of three of its four engineers. “75% of the people on our engineering team lost their jobs here yesterday because of the brutal impact AI has had on our business,” he wrote. No strategic framing, no efficiency narrative—just a small company saying plainly that AI had undercut its product’s value proposition, and the people had to go.UPS rounded out the major announcements with a plan to eliminate 30,000 jobs across its operational workforce through 2026, using attrition and voluntary separation programs, while closing 24 facilities in the first half of the year.
What the rest of 2026 look like
Tech workers, recruiters, and economists are all looking at the same uncertain horizon.The layoffs carry risks that go beyond the human cost to the individuals affected. Cutting aggressively tends to push out exactly the people who have options elsewhere—the most skilled engineers, the best product thinkers—while retaining those with fewer alternatives. It damages morale, slows product development, and has a way of creating the startup competitors that big companies fear most.There’s also a reputational problem building. The more companies frame AI as a reason to eliminate jobs, the more the technology gets associated with displacement rather than opportunity—which is already fueling local resistance to data center construction in communities across the country. That’s the last thing companies need when they’re betting the house on AI infrastructure.For now, 92,272 tech workers are out of a job, and 2026 isn’t close to done. Whether this is the industry course-correcting after years of excess, or something more structurally permanent, is a question that probably won’t get a clear answer for another year or two. What’s already clear is that for tens of thousands of people, the wait for that answer is happening without a paycheck.
Leave a Reply