110,000 layoffs, $725 billion in AI: the paradox that defines 2026

On June 22, 2026, Oracle disclosed in an annual regulatory filing that it had cut 21,000 employees over the past 12 months — 13% of the company. No prior public announcement. No press conference. Just a paragraph in the SEC disclosure: "the adoption and deployment of AI technologies across our operations have resulted, and may continue to result, in reductions to our workforce".

The next day, June 23, the Nasdaq closed 2% lower led by Micron, amid a global tech sell-off. That same morning, CNBC published that Microsoft, Amazon, Meta, and Google will spend $725 billion on AI in 2026 — more than the entire global oil and gas industry spends on exploration.

Those two numbers, seen together, tell the whole story of the tech market in 2026.

What happened in 6 months

Between January and June 2026, more than 110,000 tech workers were laid off across 137 companies, per Layoffs.fyi. To put that in perspective: 864 people per day, every day, for six months. And the curve is accelerating — May saw the highest monthly volume of tech layoffs in years.

The running list:

Meta — 8,000 cuts starting May 20. Two more rounds expected in 2026 (August and fall). Mark Zuckerberg, in 2022, apologized for over-hiring. In 2026, he simply said he doesn't know "what the optimal size of the company will be in the future".
Microsoft — 8,750 buyouts. First time in the company's 51-year history. Structured as "voluntary separation", but functionally involuntary.
Amazon — 16,000 corporate cuts in Q1, more than half of all tech layoffs for the quarter. Meanwhile, AWS grew 24% — the fastest pace in 13 quarters.
Oracle — 21,000 over 12 months. Stock down 25% since January 2026. Internally, the corporate Slack lost 10,000 users in a single day.
Intuit — 3,000 cuts (17% of headcount). CEO Sasan Goodarzi told CNBC: "none of it had to do with AI". The official press release said the opposite.
Cisco — nearly 4,000. Stock popped 13% the day after the announcement.
Coinbase — 700 cuts, org "flattened to 5 layers below the CEO". Experimenting with "one-person teams" combining engineering, design, and product.
GitLab — 350 (14% of staff) on June 3. CEO Bill Staples called it a "generational rebuild" and said agentic workloads are "pushing competitors to the brink".
GM — 500–600 IT employees in Austin and Warren.
Snap — 1,000.
Block — 4,000 cuts in March. CEO Jack Dorsey: "intelligence tools have changed what it means to build and run a company". Stock jumped the next day.

And that list only covers named, announced layoffs. Trackers like Challenger, Gray & Christmas peg 95,878 workers impacted across 249 events in the first four months alone.

The math nobody wants to do in public

The other side of the equation:

Microsoft: $190 billion in 2026 capex.
Amazon: $200 billion.
Meta: between $125 and $145 billion (raised the ceiling by $10B in May). Equals $370 million per day in data center construction.
Google/Alphabet: ~$190 billion.

Total: $725 billion in 2026, projected to hit $1 trillion by 2027.

Susan Li, Meta's CFO, said it openly on the Q1 earnings call: "we've continued to underestimate our compute needs". Translation: models keep getting larger, costs keep climbing, and the only expense flexible enough to absorb that is payroll.

Put coldly: layoffs fund GPUs.

CNBC reported Nike, in an unusual move for a non-tech company, laid off 1,400 people in May — almost all in the technology department. When companies that sell shoes start cutting IT to free up cash, the boundary of what counts as a "tech layoff" has dissolved.

Video coverage

Bloomberg covered the acceleration of tech layoffs against the rest of the market's recovery:

"AI washing" is a real part of the story

Not every layoff attributed to AI is caused by AI. That point needs to be made honestly.

Sam Altman, at a conference in February 2026, acknowledged: "there's some AI washing where people are blaming AI for layoffs that they would otherwise do".

Marc Andreessen was more direct: companies are using AI as "the silver bullet excuse: 'ah, it's AI'" — masking post-pandemic over-hiring corrections.

Deutsche Bank analysts wrote in January: "AI redundancy washing will be a significant feature of 2026". Oxford Economics concluded that firms "don't appear to be replacing workers with AI on a significant scale".

The numbers back the skepticism. Challenger, Gray & Christmas ranks AI as the 5th most common reason for cuts, behind market conditions, restructuring, closures, and cost-cutting. Andy Challenger summarized it precisely:

"Companies are shifting budgets toward AI investments at the expense of jobs. The key word is budgets — the money funding AI infrastructure is coming from payroll, even if the specific jobs being eliminated aren't being done by AI."

So: AI may not be writing the code you used to write. But it's absorbing the money that paid you to write it. For people who were let go, the distinction is purely semantic.

What the data says about devs in 2026

For people inside the industry, a few indicators matter more than headlines:

Bay Area time-to-hire: jumped from 38 days in Q3 2025 to 67 days in Q1 2026. Senior engineers cut from Salesforce, Intel, and Workday are job-searching at the highest rate since 2022.
Tech salaries: largely flat since 2025, except for AI-specialized roles (Motion Recruitment, 2026).
Junior hiring: Motion Recruitment's study shows AI adoption "is slowing hiring for entry-level and generalized IT roles".
Glassdoor Employee Confidence Index: the tech sector had the largest year-over-year drop of any industry — 6.8 percentage points in March, hitting 47.2%.
Quit rates: people are quitting less. Daniel Zhao, Glassdoor's chief economist: "because natural attrition isn't happening as much, companies are being more aggressive about pushing people out of the door".

One number provides some comfort: IBM tripled entry-level hiring in 2026 — because, per the company, AI "needs human oversight, training data curation, and judgment". At least one path is opening at the opposite extreme.

What this means if you live by code

Strip the noise away. Three things are happening simultaneously:

Companies are treating tech workers as a flexible cost line for the first time in a decade. Post-2022, Big Tech learned it can cut aggressively without market punishment — quite the opposite, the market rewards it.
The competitive advantage of "being at a Big Tech company" has dropped dramatically. Job stability is now fiction. Stock option packages are bets on companies that price you as a variable.
The barrier to building software has dropped just as much. In June 2026, a single developer with Claude Code can ship what a team of 5 shipped in 2022. That opens up potential for anyone wanting to build something — and squeezes anyone depending on a seat in a big machine.

The practical takeaway, for me, is uncomfortable to say but honest:

My read as a Brazilian dev

Being employed at a Big Tech in 2026 has become riskier than running a small software house. I'm not saying it's easy. I'm saying the equation has changed. Someone with a modern stack (Next.js, TypeScript, Postgres, AI agents) and even an embryonic product is better positioned to weather the next 3 years than someone waiting for the severance package from the next round.

The paradox is only apparent. The same tools eliminating salaried tech workers are the ones making it viable to build something solo or as a duo. The question is which side of the equation you want to be on.

Where this lands

Three possible scenarios for the next 18 months:

Optimistic: the Big 4 deliver real productivity gains via AI. Companies grow faster, rehire into different roles (AI engineers, infra, MLOps), and the transition becomes reorganization rather than net destruction. IBM is already an example of this path.
Cynical: AI capex doesn't deliver promised ROI. Companies just "absorbed payroll savings" and nothing else. When that becomes clear to analysts (likely Q2/Q3 2026), capex decelerates, but the cut jobs don't come back. Bloomberg estimates ~50% of cuts will be re-onshored at lower salaries or offshored.
Uncomfortable: this becomes the new normal. Companies operate structurally smaller, with higher margins, and the rest of the market adjusts to that new baseline. Those who fell behind the AI curve don't come back.

Can't tell which one will win. But for anyone who codes, the dominant strategy across all three scenarios is the same:

Build something of your own. Even small. Even on the side.
Cultivate genuine fluency in agentic engineering, not vibe coding. (I wrote about that difference in the vibe coding post.)
Treat Big Tech as a client, not a home. That distinction matters when the memo lands.

The industry is reorganizing in public, in real time, and the speed is faster than any previous cycle. The paradox of $725 billion invested in AI while 110,000 people are laid off won't resolve from the emotional side. It'll resolve from the math side.

The question every developer needs to ask is: which side of that math do I want to be on?