Author: Hua Lin Wu Wang
Editor: Jing Yu
Original Title: 170,000 People, This Silicon Valley Layoff Surpasses the "COVID-19 Pandemic"
The U.S. employment data for February 2026 is out, and one number left economists silent for a moment—the rate of job losses in the tech industry is now surpassing the levels seen during both the 2008 financial crisis and the 2020 pandemic.
These two time points represented the most severe economic shocks in the United States over the past two decades.
And now, the tech industry is surpassing both with its layoff numbers.
The question is, in 2008, the banks collapsed; in 2020, the pandemic caused lockdowns; so what is collapsing today in 2026?
01 The Bubble Burst, But Not a Valuation Bubble
Let's rewind to 2020–2022. The explosion in digital demand spurred by the pandemic, combined with the Federal Reserve's near-zero interest rates and cheap capital, made tech companies feel like they had suddenly struck gold, leading to疯狂扩张 (frantic expansion). Some leading companies doubled or even more than doubled their headcount in just two to three years.
The logic back then was simple—growth was the only KPI, burning cash was the only way, and headcount was the only execution tool.
Then interest rates rose. The foundation of the growth logic shook, valuations began to fall, investors became cautious, and layoffs quietly started at the end of 2022. But at that time, most people still saw it as an "adjustment," thinking everything would return once the market improved.
But it didn't return.
Throughout 2025, the global tech industry cut approximately 245,000 jobs. U.S. companies contributed nearly 70% of that, over 170,000 people.
Entering 2026, the momentum hasn't slowed; it has accelerated—in just the first six weeks, over 30,000 people have been laid off, with over 80% coming from U.S. companies.
Amazon, after recording a record $71.69 billion in revenue in 2025, announced it would cut 16,000 corporate jobs in 2026, accounting for more than half of all announced tech layoffs.
Block's CEO Jack Dorsey wrote in a letter to shareholders, "Smaller teams using the tools we are building can do more and do better." Autodesk and Salesforce each cut about 1,000 jobs at the beginning of the year.
Note this detail—most of these companies are still profitable, some even setting revenue records.
This isn't a survival layoff; this is a proactively chosen layoff.
02 Is AI the Scapegoat?
Every large-scale layoff needs a narrative to explain it.
This round, AI has become the most convenient one.
"Layoffs due to AI replacement"—this statement has a technological and contemporary feel, sounding irrefutable. But the data tells another story.
According to statistics from RationalFX, of the approximately 245,000 global tech layoffs, only about 69,800 (approximately 28.5%) can be directly attributed to the adoption of AI and automation.
That is to say, over seventy percent of the layoffs have other reasons behind them.
IBM's CEO Arvind Krishna directly addressed this issue: "From 2020 to 2023, some companies' employee numbers grew by 30% to 100%, this is just an adjustment companies need to make." He didn't blame AI, but pointed to a simpler truth—the economic hangover after over-hiring.
Of course, AI isn't completely innocent. It's just that its mode of action is more subtle than "direct replacement"—AI made companies realize that so many positions simply didn't need to exist. It didn't fire a specific person; it made management recalculate the books and find the numbers didn't add up.
This logic is more cruel and harder to refute. It's hard to tell a company "my job can't be done by AI" when it actually does it.
One analyst used a term to describe this round of layoffs—a "structural reset," not a "short-term cost correction." The difference between the two is that the latter means you might come back when the market improves, the former means that position is gone for good.
This is the most important factor in understanding this tech winter.
Past large-scale layoffs were essentially temporary contractions on the demand side. Companies were waiting for the economy to recover; once it warmed up, the same positions would reopen. But this time, many eliminated positions are being permanently redesigned—companies are rebuilding their organizational structures around AI-first workflows.
General Assembly's CEO Daniele Grassi gave a sober warning: while companies are cutting headcount and increasing AI investment, this is creating a skills gap, and this gap will ultimately slow down the transformation speed.
In other words, the layoffs themselves are creating new risks.
From market data, the tech industry shows a peculiar polarization—demand for AI-related roles is surging, while demand for traditional general tech roles is shrinking. "Tech is both growing and contracting," and these two things are happening simultaneously, just to different people.
If you are an engineer with an AI background, understand prompt engineering, and can optimize large model inference costs, the 2026 job market might be the best it's been in years for you.
If you are a general product operator, mid-platform engineer, or traditional salesperson, you might be facing a rapidly narrowing market.
This isn't the industry declining overall; it's the industry rapidly redefining "valuable people."
03 How Cold Will This Winter Be?
The judgment of Oxford Economics' chief economist Adam Slater is alarming—if the tech industry continues to decline, U.S. GDP growth in 2026 could fall to 0.8%, hovering on the edge of "near recession."
Excluding tech investment, the U.S. had almost no growth in the first half of 2025.
The U.S. economy's reliance on technology has become so deep that a move in one part affects the whole.
But there is another side to the story. A Salesforce industry observer pointed out that if you compare the absolute layoff numbers for the whole of 2025 with 2024, they actually decreased by about 20%. The narrative of "2025 being a disaster year" isn't entirely supported by the data.
This wave of layoffs is more like a transition period without a clear endpoint, rather than a decline with a bottom from which to rebound.
Companies are using layoffs to "free up space"—space for AI tools, for leaner teams, for higher human efficiency ratios. This logic will hold until some boundary is touched—maybe regulation, maybe a technological bottleneck, maybe some consumer reaction.
Jack Dorsey's phrase, "smaller teams, doing more,"某种程度上代表了整个行业此刻的集体信仰 (to some extent represents the collective faith of the entire industry at this moment). The question is, when everyone is getting smaller, who will support the next "bigger"?
What the tech industry is experiencing is not an ordinary cyclical trough, but a fundamental questioning of "what role do humans play in the system?"
Unfortunately, layoff numbers can't provide the answer to this question.
Twitter:https://twitter.com/BitpushNewsCN
BitPush TG Discussion Group:https://t.me/BitPushCommunity
BitPush TG Subscription: https://t.me/bitpush
