When a 5,000-word essay written from an imaginary future can erase hundreds of billions in market value before lunchtime, something deeper than ordinary volatility is at work. Wall Street, for all its sophisticated modeling and algorithmic firepower, may be running on a foundation of assumptions that are far more fragile than the trading floors would like to admit.
That was the uncomfortable lesson delivered in late February when Citrini Research, a Substack-based macro analysis firm that had already built a devoted following for its early calls on AI and weight-loss pharmaceuticals, published what it called a “thought exercise in financial history, from the future.” Titled “The 2028 Global Intelligence Crisis,” the essay posed a single, devastating question at the outset that the financial press largely buried under its alarm about the market reaction: What if our AI bullishness continues to be right — and what if that’s actually bearish?
The piece framed itself as a macro memo written in June 2028, looking backward at the collapse. It described an S&P 500 that had plunged 38% from its October 2026 highs, an unemployment rate printing above 10%, and an economy trapped in what Citrini called “the human intelligence displacement spiral” — a feedback loop with, in the authors’ words, “no natural brake.”
In the scenario, white-collar workers representing half of U.S. employment and driving roughly 75% of discretionary consumer spending had been systematically displaced by AI agents. The companies selling workflow automation software were themselves disrupted by better workflow automation. Their response — cut staff and double down on AI — only accelerated the spiral. Markets didn’t take kindly to the thought experiment.
When Fiction Moves Markets
The immediate market reaction told its own story. Hedge funds had already pocketed an estimated $24 billion by shorting software stocks in 2026, even before the Citrini memo went viral, as the sector bled roughly $1 trillion in market capitalization since January. But the essay’s Monday release turned a grinding sector decline into a flash panic.
Software firms DataDog, CrowdStrike, and Zscaler each dropped more than 9% in a single session. IBM suffered its worst single-day performance since 2000, falling 13%. American Express, KKR, and Blackstone — all named explicitly in the Citrini scenario as downstream casualties of white-collar job loss — tumbled alongside them. The iShares Expanded Tech-Software ETF (IGV), which tracks the sector Citrini targeted, hit a new 52-week low, falling 5% on the day and erasing all gains made since ChatGPT’s public debut in November 2022.
The VIX — Wall Street’s so-called “fear gauge” — jumped 14% and crossed above 20. Gold climbed nearly 3% above $5,200 per ounce. The flight to safety was instantaneous and measurable. And the instrument that triggered it wasn’t a Fed announcement, a geopolitical crisis, or a corporate earnings implosion. It was a Substack post.
That deserves to be said plainly, because the financial press largely focused on debating the merit of the Citrini thesis while glossing over the more alarming reality the episode exposed: that modern financial markets — leveraged, algorithmically amplified, and sentiment-driven to a degree never before seen in history — can now be materially moved by speculative narrative. Warren Buffett famously called derivatives “financial weapons of mass destruction” in his 2002 letter to Berkshire Hathaway shareholders.
What the Citrini episode demonstrated is that in an era of AI-accelerated information flow and social media virality, the weapons no longer need a detonator in the traditional sense. A well-crafted scenario distributed on the internet can function as the trigger.
The Thesis Itself — And Why It Landed
The Citrini scenario deserves a fair reading, regardless of where one lands politically or economically, because its core mechanism is not unreasonable. The essay’s central argument is that the U.S. economy is fundamentally a white-collar services economy, and that AI is not merely automating discrete tasks but is beginning to replicate entire workflows. As a result, companies that once purchased complex software suites from firms like Zendesk or ServiceNow are discovering they can rebuild that functionality in-house using AI coding tools — what the technology world has taken to calling “vibe coding.”
The bargaining power shifts. Vendors face margin compression. To preserve profitability, the vendors cut staff. And because those laid-off workers were consumers, their absence from the spending pool triggers downstream contractions in credit, housing, and consumer discretionary sectors. The spiral accelerates.
“The company that sold workflow automation was being disrupted by better workflow automation, and its response was to cut headcount and use the savings to fund the very technology disrupting it,” Citrini wrote. “What else were they supposed to do? Sit still and die slower?”
The question is brutal precisely because it is honest. This is not science fiction — it is the logic of competition applied to a sector undergoing rapid technological substitution. And it is the reason the memo resonated even among people who found its doomsday framing overwrought.
Independent researcher Laks Ganapathi had published a near-identical scenario weeks earlier, calling it the “vibecession.” Her forecast was blunt: companies would lean on AI as aggressively and as quickly as possible, and in doing so, some enterprises would simply cease to exist as going concerns. She predicted persistent unemployment alongside stubborn inflation — a stagflationary trap in which the headline economic data looks acceptable while the lived experience of the working middle class deteriorates sharply. That divergence between statistical normalcy and experiential collapse is not a new phenomenon. It is, in fact, the story of the American economy for much of the past decade.
Citadel’s Counter — And What It Gets Right and Wrong
Ken Griffin’s Citadel Securities did not remain silent. The firm published a pointed rebuttal authored by macro strategist Frank Flight, systematically challenging Citrini’s underlying assumptions. The Citadel response is worth taking seriously, and not merely because it comes from one of the most sophisticated trading operations on earth.
Citadel’s first objection is empirical: if AI is destroying software jobs, why is demand for software engineers actually up 11% year-over-year as of early 2026, according to Indeed job posting data? The Citrini scenario, Citadel argues, relies on the “recursive technology fallacy” — the assumption that a technology can expand infinitely without encountering the friction of physical, economic, or regulatory constraints. As Flight writes, displacing white-collar work at the scale Citrini envisions would require “orders of magnitude more compute intensity than the current level utilization,” and the energy demands alone would push the marginal cost of compute above the marginal cost of human labor for many tasks — thereby creating a natural economic boundary that halts the substitution before it becomes catastrophic.
The historical argument is also compelling. The internet displaced enormous numbers of workers in publishing, retail, travel, and financial services. It did not produce a deflationary depression. Instead, it lowered costs, expanded consumption, created industries that hadn’t existed before, and ultimately raised living standards for the broad population. Citadel quotes Keynes to make the point and then turns the quote against the doomsayers: Keynes underestimated the elasticity of human wants, and so, Citadel argues, does Citrini.
But Citadel’s rebuttal has its own blind spots, and an honest assessment must name them. The internet analogy holds only if one assumes that the pace of AI displacement will allow society adequate time to adapt — to retrain workers, develop new industries, and absorb the transition costs. The Citrini thesis is not primarily that AI will eventually impoverish humanity. It is that the transition could be so rapid and so concentrated in the white-collar sector that the adjustment period will itself produce cascading economic crises. Speed matters. And the speed of AI capability development in 2025 and 2026 has routinely exceeded the forecasts of even its most optimistic proponents.
What Ordinary Americans Are Already Living
Beyond the Substack debate and the Wall Street positioning, real people are already navigating the early stages of this transformation. Nicole James, a 42-year-old who built Snapchat’s content team and rose to head of content at animation studio Invisible Universe, was laid off in 2023 when the studio pivoted to become an AI operation and cut half its staff. Her story is not unique. It is becoming representative.
The Yale Budget Lab noted in late 2025 that “the broader labor market has not experienced a discernible disruption since ChatGPT’s release 33 months ago.” That is technically accurate. It is also worth noting that 33 months is a very short window for measuring the structural effects of a general-purpose technology. The broader concern is not what has already happened but what the rate of change implies about what comes next — particularly for a country whose financial markets, consumer economy, and retirement savings are all deeply intertwined.
The Real Risk No One Is Naming
The most underreported dimension of this story is the one buried in the Goldman Sachs prime brokerage data. When the Citrini memo went viral, single-stock short sales reached their highest notional level since 2016. A basket of 164 companies in software, financial services, and asset management shed $611 billion in a single week. That is not a market processing information efficiently. That is a market reacting to fear with the velocity and coordination that only algorithmic and derivatives-driven trading can produce.
The irony is almost too precise. The very tools — AI-assisted trading, algorithmic amplification, sentiment analysis scraped from social media — that Citrini warns will displace human workers are the same tools now capable of translating a speculative narrative into a trillion-dollar market event before most Americans have finished their morning coffee. Wall Street has built an AI-accelerated market and is now genuinely uncertain whether it can control the machine it has assembled.
One client told RBC derivatives strategist Amy Wu Silverman that he describes himself as a “fully invested bear” — his portfolio looks bullish because there is “career risk” in not being long AI stocks, but he is personally convinced the underlying thesis is unsound. If that posture is widespread — and the derivatives positioning suggests it may be — then American financial markets are, at their core, running on an enormous collective act of performance rather than genuine conviction. Every participant assumes the music will stop and is simply trying not to be the one left without a chair.
That is, by any reasonable definition, a bubble. Whether it pops the way Citrini’s fictional 2028 memo describes is unknowable. But the episode serves as a stark reminder that the complexity layered into modern financial systems — the derivatives, the leverage, the algorithmic feedback loops — does not reduce systemic risk. It concentrates it, accelerates it, and ultimately places it beyond the reach of any single institution or regulator to manage in real time. What the Citrini episode demonstrated, in miniature, is exactly the dynamic its authors warned about in the macro: a system so tightly coupled and so sensitively tuned that a small perturbation can cascade faster than human judgment can intervene.
The canary is still alive, as Citrini itself acknowledged in closing. But it is singing. And the coal miners of the American middle class — the accountants, the software developers, the marketing managers, the financial analysts — would do well to pay attention to the song, regardless of whether the doomsday timeline of a Substack essay proves prophetic.
