Written by: Yellow Lobster, Deep Tide TechFlow

On February 22, an article titled "2028 Global Intelligence Crisis" went viral in the financial circles. The author is the macro research institution Citrini Research, and the format of the article is a "memo from the future," assuming the timeline is June 2028, reviewing how this economic crisis triggered by AI evolved step by step into a systemic collapse.

In the article, there is a sentence, "In early 2026, the first wave of layoffs due to human intelligence being replaced began. Profit expansion, earnings exceeded expectations, stock prices hit record highs."

Four days later, this sentence was no longer a thought experiment.

On February 26, Jack Dorsey posted on X: "we're making @blocks smaller today."

Block, the fintech company that owns Square and Cash App, released its fourth quarter financial report that day. Gross profit grew 24% year-on-year, and earnings per share exceeded analyst expectations. Meanwhile, Dorsey announced layoffs of over 4,000 people, accounting for 46% of the company's total workforce.

After the announcement, Block’s stock price rose 24% after hours.

The company's profits rose by 24%, stock prices rose by 24%, and then 4,000 people received termination notices.

Citrini's "2028 Nightmare" did not wait until 2028; it has already begun its first act this Thursday.

We Are Not in Trouble

Historically, in every major layoff, the CEO's public letter follows a fixed script: The market environment is severe, strategic direction adjustment, we made a difficult decision, thanks to every colleague for their contributions.

Dorsey’s letter is different.

“We are not in trouble. Our business is strong… but something has changed. We have seen internally that smaller teams can do more and do better in conjunction with the intelligent tools we are building and using. And the capabilities of these tools are growing compound weekly.”

No mention of a market winter, the company is doing well, but you are no longer needed, this frankness is unnerving.

In past layoff narratives, there was always an implied promise that when the market improves, we will rehire. This time, Dorsey did not give that promise. What he offered was another logic: Small teams plus AI can do the same things as large teams, or even better. Since that’s the case, why do we need so many people?

Investors completely agree with this logic, casting their votes with a 24% increase in stock price.

There’s another detail that may be overlooked.

To promote an "AI-first" work culture, Dorsey previously required every employee in the company to send him an email each week, listing five things they recently accomplished. Thousands of emails poured in, and Dorsey processed them by: summarizing them with AI and then reading the summaries.

Using AI to judge who can prove they won't be replaced by AI, letting AI analyze who will be laid off, this detail is the most precise metaphor of the entire story.

A Timeline, An Acceleration

Block is not an isolated case; it is a trend that has been running for two years.

Looking back in time, the acceleration of this trajectory is dazzling.

In 2024, Klarna CEO Sebastian Siemiatkowski publicly announced that the company’s AI customer service assistant had handled work equivalent to 700 full-time employees. At that time, most people regarded this as a tech show; the CEO needed a headline-grabbing number, a story to convince investors.

In April 2025, an internal memo from Shopify CEO Tobi Lütke was leaked. The letter contained a line that was later repeatedly quoted: “Before applying for additional staff, the team must first prove that AI cannot do this task.”

In the same year, Duolingo announced an "AI-first" strategy, terminating a large number of content creation outsourcing contracts. IBM acknowledged that it replaced 8,000 human resources positions with AI; CEO Arvind Krishna openly named the department and the number of people during interviews.

Salesforce laid off 4,000 customer support positions, with CEO Marc Benioff stating: "AI can now handle about half of the company's work."

By the end of 2025, data from the U.S. employment tracking agency Challenger, Gray & Christmas showed that more than 55,000 layoffs were directly attributed to AI that year.

At the start of 2026, Amazon announced layoffs of about 30,000 corporate positions in two rounds. The law firm Baker McKenzie followed with layoffs of 600 to 1,000 research, marketing, and administrative support positions, an industry that was widely regarded as one of the most challenging for AI to penetrate.

On February 26, 2026, Block. A profitable company laid off 46% of its employees at once.

But layoffs are just the most visible knife.

A study from Harvard revealed a more insidious number: After the widespread adoption of AI, technology companies, on average, reduced hiring of junior employees by 5 each quarter. Without announcements and press releases, positions quietly disappeared from recruitment websites, and resumes sent by new graduates sank without a trace; the reasons were never written in rejection letters.

The Spiral Citrini Speaks Of

Returning to that viral article.

Citrini's reasoning is unsettling not only because it portrays a dystopia where AI sweeps the job market, but also because it depicts a logically coherent and entirely rational death spiral.

The operation of the spiral works like this:

AI expands company profits. The funds from profit expansion are reinvested into AI, more investment brings stronger AI capabilities. Stronger AI capabilities make more roles interchangeable. More unemployment means less consumer spending. The shrinkage on the consumption side puts more companies under pressure, forcing them to further reduce costs using AI. AI’s capabilities improve further.

Citrini named this cycle: Intelligence Displacement Spiral.

They wrote in the article: “Every individual decision made by a company is rational, but the collective outcome is disastrous.”

Now, in comparison to what happened at Block on this day. Gross profit increased by 24%, stock price rose by 24%, 4,000 people were unemployed, and the money saved continued to be invested in AI tools. From Dorsey’s perspective, this is a completely rational decision. He even explained in his open letter why he chose to lay off on such a large scale all at once rather than gradually: because it would undermine morale and trust continuously.

From a corporate governance perspective, this is textbook execution. From the perspective of those 4,000 individuals, this is a rupture in life.

Within Citrini's reasoning, there is a real person (presented anonymously): a friend who worked as a senior product manager at Salesforce, earning $180,000 a year, who lost his job in the third round of layoffs in 2025. He searched for six months without finding a position at the same level. Eventually, he started driving for Uber, bringing in an annual income of $45,000.

This is not just one person’s story.

Citrini performed a simple multiplication in the article: taking this individual trajectory and multiplying it by the hundreds of thousands of white-collar workers in each major city experiencing similar fates, the shrinkage on the consumption side is no longer an abstract macro data but a foreseeable and calculable reality.

This story is synchronously playing out around the world, perhaps right next to you and me.

No Bad Guys to Be Found

The article by Citrini states:
“The historical disruption model suggests that incumbent enterprises will resist new technologies and will be gradually eroded by more flexible newcomers, leading to their decline. The experiences of Kodak, Blockbuster, and Blackberry are just like that. But the situation in 2026 is completely different; the reason existing enterprises do not resist is that they cannot afford the cost of resistance.”

This is the key to understanding the entire situation.

Klarna has been impacted by AI, then compresses costs with AI and laid off a group of people. Salesforce’s software products were challenged by AI, then replaced 4,000 customer support positions with AI. Block was impacted by the wave of AI in the fintech industry, then announced a restructuring of the entire organization using AI, laying off nearly half of its employees.

They are not victims defeated by AI. They are AI’s most active adopters, and what is defeated is their own employees.

This is the part that is hardest to handle within a moral framework.

After the 2008 financial crisis, people knew who to hate. The bankers on Wall Street, the traders packaging and selling junk bonds, the regulators who were absent. Anger had specific targets, even addresses, hence the Occupy Wall Street movement.

This time is different.

It is difficult to say that Dorsey did anything wrong; Block’s stock price tells you what the market thinks, and the 4,000 laid-off people did nothing wrong either; they just happened to be working in roles that are being restructured. AI itself is certainly not the bad guy; it is merely a tool that is becoming increasingly useful at a pace never seen by humanity.

Responsibility is scattered throughout the system, like salt dissolving in water; you can taste the saltiness, but you cannot find that grain of salt.

There are two sentences in Citrini's article that have not been widely quoted, but perhaps have the deepest meaning in the entire text:

“This is the first time in history that the most productive asset in the economy has created fewer jobs rather than more. No framework applies because they were all not designed for such a world, where scarce production factors have become abundant.”

After every technological revolution in the past, humanity has found new positions. The steam engine replaced manual textile workers, but created railroad workers, factory managers, and urban planners. The internet wiped out travel agencies, physical record stores, and classified ads, but invented product managers, data analysts, and content creators. Every time, those "jobs of the future" could not be specifically described at first, but they appeared later in sufficient numbers.

This comforting pattern has encountered a challenger for the first time.

Because this time, those "jobs of the future," like AI trainers, prompt engineers, and AI product managers, AI itself is also learning to do. The displaced workers cannot simply "upgrade their skills" to transition to AI-related positions because that position itself is also being compressed.

Harvard researchers recorded a phenomenon: After the widespread adoption of AI, hiring for junior positions in tech companies has decreased by over 50%. Not because these positions have disappeared, but because they have not been created at all.

A whole generation trained to enter an industry then quietly decided it no longer needed entry-level humans just as they were about to graduate.

We cannot pretend that we still have time to think it through slowly.

Citrini concludes that the canary is still alive, but the miner's problem has never been whether the canary is dead or alive; it’s whether you have an exit when it starts to wobble.