Tech sector job cuts surge 24% in March 2026 as AI spending rises

Technology companies cut 18,720 jobs in March 2026, a 24% increase compared to March 2025, according to outplacement firm Challenger, Gray and Christmas. The number is not a one-month anomaly. It fits a pattern that has been building since late 2024, where large tech employers reduce headcount in specific functions while spending more on AI infrastructure, data centers, and automation tooling.

The layoffs are happening at profitable companies. That is what makes this cycle different from the 2022 and 2023 cuts, which followed a period of overhiring during the pandemic and a sharp correction in ad revenue and consumer spending. This time, the companies announcing cuts are often posting strong earnings. The job reductions are budget reallocations, not emergency measures.

Where the cuts are landing

The roles being eliminated are concentrated in middle management, customer support, content moderation, and certain categories of software development that AI tools are beginning to automate. Oracle announced cuts of an estimated 20,000 to 30,000 workers in the US and India in early 2026 while simultaneously increasing its data center investment budget. That pattern, fewer people and more compute, is showing up across enterprise software, cloud services, and consumer platforms.

AI coding tools have changed the math on engineering headcount. GitHub Copilot, which had over 1.8 million paid subscribers as of early 2025, lets individual developers write and review code faster than before. When a team of 10 engineers can now produce what previously required 14, companies doing annual headcount planning notice. They do not always announce it as an AI-driven reduction, but the connection is visible in the numbers.

AI spending is going up as payroll goes down

The capital being freed from payroll reductions is not sitting idle. Microsoft, Google, Amazon, and Meta collectively announced over $300 billion in combined AI and data center investment plans for 2025 and 2026. That spending goes toward Nvidia GPUs, power infrastructure, networking equipment, and cooling systems. None of that requires a large permanent workforce in the traditional sense. A data center with 50 megawatts of compute capacity can be operated by a few hundred people.

The shift creates a structural mismatch in the labor market. The people being laid off from support, moderation, and mid-level management roles do not automatically qualify for the data center construction jobs, AI safety roles, or ML engineering positions being opened. Retraining helps at the margins, but the skill gap between a terminated content moderator and a newly opened AI trainer or infrastructure engineer is not something a six-month bootcamp fully closes.

How this compares to previous tech layoff cycles

The 2022 and 2023 tech layoffs were the most covered in recent memory. Companies like Meta, Amazon, and Google cut tens of thousands of positions after interest rate increases deflated growth-stock valuations and advertising budgets tightened. Those cuts eventually slowed, and hiring resumed in 2024, though at more selective levels. The 2026 cuts feel different because they are arriving while revenue is growing, not contracting.

Challenger, Gray and Christmas tracked 72,821 tech job cuts in the first quarter of 2026. That puts Q1 2026 ahead of Q1 2025 in total tech sector cuts, despite the overall economy continuing to grow. The Bureau of Labor Statistics reported US unemployment at 4.1% in February 2026, which means most of these workers are finding jobs eventually. But the re-employment timeline and salary outcomes vary significantly depending on the role and skill set.

What workers in at-risk roles should be thinking about

The functions most exposed to continued cuts are those where AI tools are already producing measurable output at lower cost per unit than human labor. That list currently includes first-line customer support, document summarization, basic QA testing, and routine data analysis. Workers in these areas who have not started building adjacent skills are likely to find the job market increasingly thin over the next 18 to 24 months.

That does not mean every role in those categories disappears. But the number of open positions will compress as AI handles a larger share of the volume. Companies are already managing higher ticket loads with fewer agents by routing simpler queries to AI and escalating only complex cases to humans. Zendesk reported in its 2025 Customer Experience Trends Report that AI resolved 80% of customer interactions without human involvement at companies that had fully deployed its AI tools.

The March 2026 figure of 18,720 cuts will likely be followed by similar or higher monthly numbers through Q2, based on the pace of announced restructuring plans at Oracle, Intel, and several mid-size SaaS companies that have not yet completed their workforce adjustments.