Citrini's 'thought exercise' on AI sparks selloff in Visa, DoorDash, others

A viral thought exercise by Citrini Research has sparked a sharp selloff in the stocks of several companies mentioned in the model scenario.

The report, titled “The 2028 Global Intelligence Crisis,” lays out a hypothetical future in which rapid advances in AI trigger a profound repricing of human labour, leading to white-collar obsolescence and a systemic consumption collapse.

Despite Citrini Research explicitly prefacing the report as a "scenario" rather than a "prediction"—dismissing it as neither "bear porn" nor "AI doomer fan-fiction"—the market reaction was swift.

The specific mention of established companies to illustrate how their respective industries might be hollowed out by AI proliferation proved too vivid for investors to ignore, sparking a defensive selloff across the highlighted names.

Shares of Visa, Mastercard, ServiceNow, DoorDash and Blackstone all fell on Monday after the report gained traction on social media.

By early afternoon in New York, Visa was down nearly 4.5%, Mastercard had slid 6.3%, ServiceNow was lower by about 4%, and DoorDash had dropped roughly 7%.

A scenario, not a forecast

Citrini’s authors were careful to frame the document as a thought experiment.

At the outset, they described it as neither a prediction nor an attempt at sensationalism, emphasising that it was meant to explore how economic systems might behave if machine intelligence became abundant and cheap.

The report was co-authored by Alap Shah and centres on the idea that human intelligence, historically the scarcest and most valuable input in the global economy, is undergoing a rapid and disruptive repricing.

As AI systems become capable substitutes for complex white-collar work, the authors argue, the “intelligence premium” that underpinned middle-class wages and corporate pricing power begins to erode.

In the scenario, this process triggers what the report calls an “intelligence displacement spiral”, a negative feedback loop in which job losses, weaker demand and falling valuations reinforce one another.

Cassandra Unchained, the official account for hedge fund manager Michael Burry, also referred to the Substack article in a post on X.

A bleak hypothetical future and likely emergence of 'Ghost GDP'

To make its argument concrete, the report adopts the format of fictional future news dispatches.

One such vignette imagines June 2028, when the global economy is grappling with rising unemployment and slowing growth.

In that hypothetical world, the unemployment rate surprises to the upside at 10.2%, triggering a 38% drawdown in the S&P 500 from its October 2026 peak.

The report argues that even as headline economic activity weakens, official data struggle to capture what it terms “Ghost GDP” — productive output generated by machines that no longer translates into wages or broad-based purchasing power.

These imagined data points were not presented as forecasts, but as narrative devices designed to stress-test current assumptions about growth, profits and labour markets.

ServiceNow and the SaaS squeeze

One of the most detailed scenarios focuses on enterprise software.

In a fictional October 2026 news item, the report describes ServiceNow’s annual contract value growth slowing sharply to 14% from 23%, followed by a 15% workforce reduction and a steep share price fall.

The model suggests that AI-driven coding tools collapse barriers to entry, intensify competition and erode pricing power across software-as-a-service.

As customers cut headcount using AI, licence-based revenue models come under pressure, creating a feedback loop in which efficiency gains at clients mechanically reduce vendors’ own sales.

DoorDash and the erosion of habit

Similarly, to illustrate how machine intelligence could dismantle "habitual intermediation," the Citrini report centers on a stark deconstruction of the food delivery giant DoorDash.

The report argues that the arrival of sophisticated coding agents has effectively demolished the barrier to entry for the delivery sector.

According to the model, a single competent developer can now deploy a functional competitor in mere weeks.

This has led to a hyper-fragmented market where dozens of new platforms entice drivers by passing through 90% to 95% of delivery fees—a move that collapses margins for incumbents like DoorDash and Uber Eats.

The destruction, however, is two-sided.

While AI enables new competitors, it also changes the nature of the consumer.

The report notes that the traditional DoorDash moat was built on a simple human premise: "you’re hungry, you’re lazy, this is the app on your home screen."

In contrast, an AI agent does not have a "home screen" or a sense of brand loyalty.

These digital assistants continuously scan DoorDash, Uber Eats, direct restaurant sites, and emerging niche alternatives to secure the lowest fee and fastest delivery time for every single order.

As the report concludes, the "habitual app loyalty" that served as the bedrock of the modern gig-economy business model simply ceases to exist when the machine takes over the transaction.