Technology layoffs are not slowing because Big Tech ran out of money. Challenger, Gray and Christmas says the sector announced 139,156 U.S. job cuts in the first half of 2026, while the biggest AI spenders are still planning hundreds of billions of dollars in data-center spending.
There is a number worth sitting with: 83%. According to Challenger, Gray and Christmas, technology companies announced 139,156 U.S. job cuts through June 2026, up from 76,214 in the same period last year. That is not a hiring pause. It is a sharp increase in the pace at which tech workers are being pushed out.
The more uncomfortable part is where the money is going. Bloomberg reported in April that Amazon, Alphabet, Microsoft and Meta were planning as much as $725 billion in capital expenditure this year, primarily for AI data-center equipment. The Financial Times reported the same broad pattern today, with U.S. tech groups cutting nearly 140,000 jobs in 2026 even as AI spending keeps climbing. You do not need a theory to understand that. You need the filings and the layoff notices.
Oracle gives you the cleanest version of the story. Reuters reported in June that Oracle's workforce fell by about 21,000 employees in fiscal 2026, from roughly 162,000 to 141,000. In its annual filing, the company said the "adoption and deployment of AI technologies across our operations have resulted, and may continue to result, in reductions to our workforce." That sentence is doing a lot of work. It says the quiet part in regulatory language.
The math is not complicated. Payroll is being converted into compute. Money that once supported teams is being moved toward chips, servers, data centers and the people who can make those systems useful. When a profitable company cuts staff while increasing AI infrastructure spending, that is not a paradox. It is a capital allocation decision.
The jobs are not moving evenly
The first roles under pressure are the ones AI systems can absorb most easily: routine support, internal operations, moderation, testing and repetitive administrative work. The companies can call this efficiency, restructuring or modernization. Fine. The effect on the worker is the same.
IBM is a useful counterexample because it shows the split clearly. Bloomberg reported in February that IBM planned to triple U.S. entry-level hiring in 2026, with roles rewritten for an AI-heavy workplace. IBM's own writing on the shift says entry-level work is moving away from purely task-driven duties and toward analysis, customer judgment and better decisions about where AI should and shouldn't be trusted. That sounds constructive. It also tells you who gets left behind.
A person cut from a support desk is not automatically hired into an AI operations role. A QA tester is not instantly a model evaluator. A junior HR worker whose tasks are handed to an agent does not become a cloud architect because the company also has cloud openings. This is the gap the industry keeps skating past. New jobs can exist and still be inaccessible to the people losing the old ones.
Sam Altman put a necessary caveat on the whole discussion. In a CNBC-TV18 interview at the India AI Impact Summit, later reported by Fortune and the San Francisco Chronicle, he said there is "some AI washing" where companies blame AI for layoffs they would have made anyway, alongside real displacement from AI. He is right to separate the excuse from the cause. Frankly, that does not rescue the sector's argument.
Challenger's June report says AI was cited in 101,743 U.S. job-cut announcements so far this year, about 23% of all cuts tracked across the economy. In June alone, AI accounted for 14,029 announced cuts, or 31% of the month's total. Those are employer-stated reasons, not proof that a chatbot personally replaced every worker. Still, the pattern is now measurable. It is not just conference-stage anxiety.
The old defense is wearing thin
For years, the technology industry's answer to displacement fears was simple: technology creates more jobs than it destroys. The printing press, the factory, the internet. Each wave removed categories of work and produced new ones at scale. Regulators accepted that story because, for long stretches, it was true enough.
That defense is harder to make when the same companies are using AI in layoff explanations while pushing capex plans past $700 billion. Microsoft said on July 6 that it was eliminating around 4,800 roles, about 2.1% of its global workforce, as it focused people and investments on its highest priorities. The company also told investors in its latest earnings materials that calendar-year 2026 capital expenditure would be roughly $190 billion, going into cloud infrastructure, GPUs, CPUs and the data centres to house them.
Look at the language. The cuts are about focus and pace - moving faster, getting leaner. The spending is about capacity and AI infrastructure. Both can be true. Together, they tell you which kind of worker the sector values more right now.
The risk is not that AI destroys every job. That is the lazy version of the debate. The risk is that the gains arrive quickly for shareholders and infrastructure vendors while the losses land immediately on workers whose next job requires credentials or experience they do not have - and who live nowhere near where those new roles are being created. IBM tripling entry-level hiring is real. Oracle shrinking by 21,000 workers is real too.
The next test is not whether Big Tech can spend more. It clearly can. The test is whether the companies using AI to justify leaner workforces can show, in numbers rather than slogans, that the people displaced by this shift have somewhere plausible to go.
Also read: The UK and Israel just appointed their first national AI chiefs and every founder should pay attention • Waymo fires Uber as a partner and the robotaxi war starts for real • Morgan Stanley says SpaceX at $100 prices its entire AI business at zero
