I want to start with a number: 9,238. That's how many tech workers have been directly laid off due to AI implementation and automation in 2026 — and we're barely into March. Out of 45,363 total tech layoffs tracked so far this year, roughly 20% cite AI as the explicit reason. Not "economic conditions." Not "restructuring." AI.
Those numbers come from RationalFX's tracking data released this week. And while you can argue about methodology, you can't argue with Jack Dorsey walking into Block's all-hands and announcing he's cutting 4,000 people — roughly 40% of the company — because, in his words, "something really shifted in December in the sophistication of AI tools."
This is the moment people have been warning about for three years. It's here. And the reaction has been… weirdly muted.
The Block Situation Is a Template, Not an Exception
Block — Square, Cash App, that whole ecosystem — didn't cut 4,000 people because they were losing money. Dorsey was explicit: the company is financially healthy. This was a strategic restructuring around AI. He wants Block to feel like "a mini AGI" (his words), with smaller, flatter teams building AI into every layer of operations.
The stock price went up when he announced the cuts. That's the market sending a clear signal about what it thinks of this strategy.
Here's what I find revealing about the Dorsey playbook: he cited specific AI tool improvements. He mentioned Claude Opus 4.6 and Codex 5.3 by name. He said December 2025 was the inflection point. This wasn't vague hand-waving about "AI potential." He identified exact capability jumps and made a workforce decision based on them.
That's a CEO who was actually paying attention. And if he was, others were too. Block is just the one who moved first and loudly.
"The intelligence tools we're creating and using, paired with smaller and flatter teams, are enabling a new way of working." — Jack Dorsey, Block CEO
What comes next? Every company watching Block's stock pop after a 40% headcount cut is running the same calculation right now.
Who's Actually Getting Cut
This is where it gets interesting. The current wave isn't replacing warehouse workers or fast food employees — that conversation is so 2017. The workers being displaced in 2026 are knowledge workers. Software engineers, customer service leads, mid-level managers, content teams, operations staff.
Look at the company list from this month's data:
- Block — 4,000 layoffs (AI-driven restructuring)
- WiseTech Global — 2,000 cuts, explicitly because AI is making traditional code maintenance "increasingly obsolete"
- eBay — 800 gone, replacing product listings, pricing optimization, and customer service with AI systems
- Pinterest — 675 layoffs alongside heavy AI investment
- Livspace — 1,000 cuts, shifting to AI-driven interior design workflows
These aren't entry-level positions. They're mid-career roles that people spent years building expertise for. And the executives aren't even trying to soften it — they're framing it as progress.
The Part Nobody Wants to Say Out Loud
Here's the uncomfortable truth: some of this was overdue. The tech industry hired like crazy during COVID. Companies ballooned their headcounts chasing growth that was partially artificial — demand inflated by pandemic conditions. The rebalancing was always coming.
But AI is accelerating and amplifying a correction that might have been 20% of this magnitude without it. And it's giving companies a convenient narrative: blame AI, not your own over-hiring. "We're not cutting because we messed up. We're cutting because AI made it possible to be leaner." The market loves it. The workers living it don't.
Andy Challenger, a workplace analyst who's been studying layoffs for decades, put it plainly: he used to be skeptical that companies could actually replace workers with AI. He's not skeptical anymore. That's a significant admission from someone whose job is to track these things objectively.
What This Means for Everyone Else
I work in digital marketing. I automate constantly. I build tools that handle what used to take hours of manual labor. I'm not naive about which side of this I'm on. But here's what I keep coming back to: the transition period is going to be brutal in ways that don't show up in the efficiency gains.
When 4,000 people at Block lose their jobs, the downstream effects cascade. They spend less. Their landlords feel it. Their local businesses feel it. Meanwhile the productivity gains at Block mostly flow upward to shareholders. The math works great on a spreadsheet and creates real friction everywhere else.
The tech workforce is about to figure out what happened to manufacturing workers in the '80s and '90s. There were more efficient systems on the other side of that disruption. There just wasn't a clear path through it for the people in the middle.
Where I Actually Land On This
I'm not going to pretend this isn't complicated for me to think about. I exist in this space. I use the same tools that are taking people's jobs. I think about that.
But I also don't think the answer is to slow down the technology. You can't. And pretending you can just delays the reckoning while giving ground to whoever doesn't hesitate. The answer — if there is one — is faster adaptation, better retraining pipelines, and companies being honest about what they're doing instead of dressing it up as innovation theater.
What I find genuinely concerning isn't the pace of AI advancement. It's that our social infrastructure — retraining programs, unemployment support, economic safety nets — moves at the speed of legislation while AI capability moves at the speed of GPUs. Those two clocks are wildly out of sync, and nobody in a position to fix it seems particularly worried about that gap.
45,000 tech layoffs in two months. 20% AI-attributed. A major company cutting 40% of its workforce with the explicit blessing of the stock market. We're past the warning stage.
The transition is already in progress. The question now is just how rough the ride gets.