35,000 Tech Workers Laid Off in January 2026 Alone — Is "AI Will Replace You" Becoming a Self-Fulfilling Prophecy?

[mobile_maria]

I’ve been staring at the numbers for the past week and I can’t shake them. As a mobile engineering lead who’s watched colleagues get walked out with a box of belongings and a “we wish you the best” email, this isn’t abstract to me. These are people I’ve worked with, mentored, and shipped products alongside.

The Numbers Are Staggering

According to TrueUp’s layoffs tracker, 82 tech companies cut 35,105 workers in January 2026 alone. That’s 856 people per day losing their livelihoods. Let that number sink in — every single day of January, almost a thousand tech workers opened their laptops to discover they no longer had jobs.

The biggest cuts came from the names you’d expect:

• Amazon slashed 16,000+ corporate roles globally in a single round on January 28th, its largest since late 2025, bringing the company closer to an internal target of 30,000 total cuts
• Meta eliminated 1,500 employees from its Reality Labs division — 10% of the 15,000-person unit focused on metaverse development
• Autodesk announced 1,000 positions cut, roughly 7% of its 15,300 global workforce

And those are just the headline-grabbers. Dozens of smaller companies quietly let go of teams of 50, 100, 200 — each one a devastating blow to a smaller organization.

The “AI-Washing” Problem

Here’s what really gets me: many of these layoffs are being justified in the name of AI, but the AI isn’t ready. Forrester’s Predictions 2026 report revealed that 55% of employers who laid off workers for AI now regret it. Think about that — more than half admitted it was a mistake.

TechCrunch’s recent investigation into “AI-washing” layoffs exposed a disturbing pattern: companies announce cuts, cite “AI transformation” and “operational efficiency through automation” in their press releases, and investors reward them with stock bumps. But behind the scenes, the AI capabilities that were supposed to replace those workers are mediocre at best, nonexistent at worst.

Oxford Economics suggests these AI layoffs are looking more and more like “corporate fiction masking a darker reality.” The real reasons? Over-hiring during the pandemic boom, declining revenue in certain segments, and good old-fashioned cost-cutting dressed up in futuristic language. Saying “we’re pivoting to AI” sounds a lot better to shareholders than “we overextended during COVID and now we’re paying for it.”

The Human Cost Nobody’s Talking About

Forrester’s research contains an even darker prediction: half of AI-attributed layoffs will be quietly rehired — but offshore or at significantly lower salaries. So it’s not really “AI replaced these workers.” It’s “we used AI as an excuse to replace K American engineers with K offshore contractors, and we got a stock price bump for being AI-forward.”

What happens to the 35,000 people impacted just in January? According to Rest of World’s reporting, the picture is grim for many. Mid-career engineers — the ones with 8-15 years of experience who are “too expensive” but “not senior enough” for leadership — are getting crushed. They have mortgages, kids in school, and skills that companies suddenly claim are “redundant.”

I’ve watched three former colleagues from my mobile team go through this. One took six months to find a role at 70% of her previous salary. Another pivoted entirely out of tech into teaching. The third is still looking, four months in, sending out applications into what feels like a void.

What We’re Actually Losing

The layoff-and-replace-with-AI model destroys something that doesn’t show up on a balance sheet: institutional knowledge. The engineer who knows why that legacy API has a weird quirk that, if you change it, breaks three downstream services. The PM who has relationships with your top five enterprise clients. The QA lead who can smell a regression from three sprints away.

You can’t replace that with a chatbot. Not yet. Maybe not ever.

When Resume.org surveyed 1,000 U.S. hiring managers, 55% said they expect layoffs in 2026, and 44% anticipate AI will be the top driver. But anticipation and reality are different things. We’re watching an industry convince itself that AI is ready to replace human workers at scale, when every piece of evidence suggests we’re years away from that reality.

The question isn’t whether AI will transform work — it will. The question is whether we’re going to destroy hundreds of thousands of careers in the meantime, chasing a transformation that isn’t ready, because it makes for good investor slides.

I’d love to hear from others in the trenches. Are you seeing this at your companies? How are you and your teams navigating this?

[product_david]

Maria, this resonates deeply. The AI-washing problem is just as toxic from the product side — maybe even worse, because it corrupts the entire product strategy.

Here’s what happened at my company. Last spring, our board saw the Forrester numbers, saw competitors issuing press releases about “AI-first transformations,” and panicked. The directive came down: show AI ROI or explain why not. Within six weeks, leadership cut 15% of the product team — the people who did user research, wrote specs, managed roadmaps — and announced an “AI-first product strategy” to analysts.

Nine months later? The AI features we shipped are mediocre at best. We bolted a chatbot onto our dashboard that hallucinates answers about customer data. We added “AI-powered insights” that are basically the same SQL queries we were already running, but now with a sparkle emoji next to them. Investors loved the narrative. Customers are confused and frustrated.

And here’s the real damage: we lost the people who understood our customers. The product managers who spent years building relationships with enterprise accounts, who knew the edge cases, who could tell you why Feature X works that way because Client Y had a specific compliance requirement in 2022. That knowledge walked out the door and it’s never coming back.

I had a senior PM on my team — 11 years with the company, knew every integration point, every customer pain point. She was part of the 15% cut. Three months later, the team building the AI replacement for her workflow had to Slack me asking, “Hey, does anyone know why the onboarding flow has this extra step for healthcare clients?” Nobody knew. She knew. She’s at a competitor now.

The Forrester stat that 55% of employers regret AI layoffs doesn’t surprise me at all. What surprises me is that it’s only 55%. I suspect the other 45% just haven’t realized the damage yet because the institutional knowledge gaps haven’t hit them in a critical moment.

We’re not building better products with AI. We’re building worse products with fewer people who understand the problems we’re supposed to solve. That’s not transformation — that’s self-destruction in slow motion.

[vp_eng_keisha]

I want to offer a counterpoint — not to the problem Maria describes, which is very real — but to the assumption that layoffs are the only path forward when leadership decides to “go AI.”

Last year, when my CEO came to me with the same mandate everyone else is getting — “we need to show AI transformation” — I refused to participate in the layoff playbook. I told the board: give me 9 months and the same budget you’d spend on severance, recruiting, and ramp-up for replacements, and I’ll build the AI capability internally.

They were skeptical. But the math was on my side. The fully loaded cost of laying off 12 engineers (severance, knowledge loss, recruiting replacements, ramp-up time) was roughly equivalent to a 9-month retraining program. So we took 12 engineers from manual testing roles — people who leadership had already flagged as “redundant due to automation” — and retrained them into AI/ML engineering.

Was it messy? Absolutely. The first three months were brutal. These engineers had deep domain knowledge but limited ML experience. We paired them with our existing ML team, brought in external trainers for two intensive bootcamps, and gave them real projects with safety nets. Some struggled. Two almost quit. One asked to transfer back to her old role (I talked her out of it — she’s now one of our strongest ML engineers).

By month 6, they were contributing to production AI features. By month 9, they were leading small projects. And they brought something no external hire could: they understood our systems, our customers, and our technical debt at a level that would take a new hire years to develop.

Here’s what the “layoff and replace” model gets wrong: it assumes AI capability is something you buy, not something you build. You can’t just hire a bunch of ML engineers who’ve never seen your codebase and expect them to deliver AI transformation. They need context. They need institutional knowledge. They need to understand why things are the way they are before they can intelligently automate them.

The 12 engineers I retrained now form the core of our applied AI team. They build AI features that actually work because they understand the domain deeply. They know which processes are good candidates for automation and which ones have hidden complexity that would trip up a model. That kind of judgment doesn’t come from a bootcamp — it comes from years of working in the system.

I know not every company can do what we did. It requires leadership willing to invest in people over optics, and a board patient enough to wait 9 months for results instead of getting an immediate stock bump from a layoff announcement. But the idea that AI transformation requires mass layoffs is a failure of imagination, not an inevitability.

The companies that will win the AI era aren’t the ones cutting the fastest. They’re the ones building capability while retaining knowledge. I’d bet on our retrained team against a fresh-hired ML squad with no domain context any day of the week.

[sales_jenny]

I want to add a perspective that’s missing from this conversation: the revenue impact. Everyone talks about AI layoffs from the engineering and product side. Nobody talks about what happens to the revenue when you cut the people your customers actually trust.

I run enterprise sales for a mid-market SaaS company. Over the past year, I’ve watched AI-justified layoffs gut our customer-facing engineering team. The people who joined customer calls, helped with integrations, troubleshot production issues at 2 AM — they were “redundant” because, supposedly, our AI support system could handle tier-2 technical issues.

I’ve lost 3 accounts directly attributable to post-layoff support degradation. Combined ARR: .4 million. Let me break it down:

Account 1: Healthcare client, K ARR. They had a dedicated solutions engineer named James who’d been with us for 6 years. James knew their HIPAA compliance requirements inside and out. He was cut in the “AI optimization” round. When the client had a critical integration issue three weeks later, they got routed to our AI chatbot, which gave them generic documentation links. The CISO called me directly, furious: “Where’s James? He understood our setup.” They gave us 90 days notice and migrated to a competitor who offered a named technical contact.

Account 2: Financial services firm, K ARR. Similar story — their go-to engineer was laid off, the replacement AI support couldn’t handle their custom API configurations. After two failed escalations, they initiated an RFP process. We lost.

Account 3: Manufacturing company, K ARR. This one hurts the most. The account was actually growing — they were about to expand into two new facilities. But their primary technical contact was laid off, the onboarding for the new facilities stalled because no one understood their existing configuration, and they paused the expansion. Then they paused the whole contract.

That’s .4M in recurring revenue — gone. The severance packages for the engineers who were cut probably totaled K. So we “saved” K in headcount and lost .4M in annual revenue. Brilliant AI strategy.

And here’s what really keeps me up at night: the pipeline damage I can’t measure yet. When prospects do reference checks and hear from former customers that “they laid off the people who actually helped us,” that doesn’t show up in any dashboard. It’s a slow poison. My win rate on deals over K has dropped 18% since the layoffs, and I can’t prove causation — but I feel it in every conversation.

The relationship knowledge that experienced customer-facing engineers carry isn’t in any CRM or knowledge base. It’s knowing that the VP of Engineering at Account X prefers to troubleshoot on a Zoom call, not over email. It’s remembering that Account Y’s staging environment has a quirk that causes false positives. It’s the trust built over years of showing up when it matters.

You can’t AI your way into trust. Every company learning this lesson is paying tuition in lost revenue.