Author｜Hualin Dance King

Editor｜Jing Yu

The US employment data for February 2026 has been released, and one figure left economists silent for a moment— the rate of job losses in the tech industry is now surpassing the levels during the 2008 financial crisis and the 2020 pandemic period.

These two time points represent the most severe shocks to the US economy in the past twenty years.

And now, the tech industry is trampling over them with its layoff numbers.

The question is, in 2008, the banks collapsed, and in 2020, the pandemic locked everything down, so what has collapsed today, in 2026?

01 The Bubble Bursts, But It’s Not a Valuation Bubble

Let’s rewind to the period from 2020 to 2022. The digital demand spurred by the pandemic exploded, combined with the Federal Reserve’s nearly zero-interest rates providing cheap capital, tech companies seemed to have suddenly discovered a gold mine, leading to frenzied expansion. The number of employees in some leading companies doubled or more within two to three years.

The logic at that time was very simple—growth was the only KPI, burning cash was the only method, and headcount was the only execution tool.

Then interest rates went up. The foundation of the growth logic became shaky, valuations began to decline, investors grew cautious, and layoffs quietly started at the end of 2022. But at that time, most people still thought this was just an "adjustment," waiting for the market to improve, and everything would return.

However, it did not return.

In 2025, the global tech industry cut about 245,000 jobs, with US companies contributing nearly 70%, more than 170,000 individuals.

Entering 2026, the momentum not only did not slow down but instead accelerated—within just the first six weeks, more than 30,000 people were laid off, with over 80% coming from US companies.

After reporting a record revenue of $71.69 billion in 2025, Amazon announced it would cut 16,000 corporate positions 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 laid off about 1,000 people at the beginning of the year.

Note this detail—most of these companies are still profitable, and some even set revenue records.

These are not layoffs for survival, but layoffs by choice.

02 Is AI the Scapegoat?

Every large-scale layoff requires a narrative to explain it.

In this round, AI became the easiest scapegoat.

“Laid off due to AI replacement”—this statement has a technical feel and a sense of the times, and it sounds indisputable. But the data tells another story.

According to statistics from RationalFX, of the approximately 245,000 tech layoffs worldwide, only about 69,800 (approximately 28.5%) can be directly attributed to the adoption of AI and automation.

In other words, more than seventy percent of layoffs are due to other reasons.

IBM’s CEO Arvind Krishna pointed out this issue directly: “From 2020 to 2023, some companies saw employee growth of 30% to 100%, this is just the adjustment that the company needs to make.” He did not blame AI but pointed to a more straightforward truth—an economic hangover after over-hiring.

Of course, AI is not completely innocent. Its role is more subtle than “direct replacement”—AI has made companies realize that many positions do not need to exist at all. It hasn’t fired anyone; it has led management to reassess the books and discover discrepancies.

This logic is more ruthless and harder to rebut. It is difficult to tell companies, “My job cannot be done by AI,” when it actually can.

One analyst used the term "structural reset" to describe this round of layoffs, rather than "short-term cost correction." The difference is that the latter implies that you will return when the market improves, while the former means that the position will no longer exist.

This is the most important factor in understanding this round of the tech winter.

The last few large-scale layoffs were essentially temporary contractions on the demand side. Companies were waiting for economic recovery; once the economy warmed up, the same positions would reopen. But this time, many of the eliminated positions have been permanently redesigned—around AI-prioritized workflows, companies are reconstructing their organizational structures.

General Assembly’s CEO Daniele Grassi provided a sobering warning: while companies cut headcount, they increased AI investment, which has created 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 positions is surging, while traditional general tech jobs are shrinking. “Technology is both growing and contracting,” and these two phenomena are happening simultaneously, only affecting different people.

If you are an engineer with an AI background, understand prompt engineering, and can optimize large model inference costs, the job market in 2026 may be the best it has been in years for you.

If you are a general product operations, middle platform engineer, or traditional sales, you might be facing a rapidly narrowing market.

This is not the overall decline of the industry, but rather the industry is quickly redefining “valuable people.”

03 How Cold Will This Winter Be?

Adam Slater, chief economist at Oxford Economics, gives a wake-up call—if the tech industry continues to decline, the US GDP growth in 2026 could drop to 0.8%, lingering on the edge of “near recession.”

Excluding tech investment, the US experienced almost no growth in the first half of 2025.

The US economy’s dependence on technology has reached a level where a single move can affect the whole body.

But there is also another voice. Observers in Salesforce noted that if we compare the absolute layoff numbers for the full year of 2025 with those from 2024, they actually decreased by about 20%. The narrative of “2025 being a disastrous year” does not completely hold up in the data.

This wave of layoffs feels more like a transition period with no clear endpoint, rather than a decline with a floor for rebound.

Companies are using layoffs to “create space,” space for AI tools, for more streamlined teams, and for higher productivity efficiency. This logic will hold until a boundary is reached—perhaps regulatory, perhaps a technological bottleneck, or perhaps a reaction from consumers.

Jack Dorsey’s statement about “smaller teams doing more” somewhat represents the collective belief 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 a normal cyclical downturn, but a fundamental questioning of “what role people play in the system.”

Unfortunately, this question, the layoff numbers cannot provide an answer.