Ken Griffin's Citadel Securities Debunks Citrini's Terrifying 2028 AI Doomsday Prediction

Citrini Research's recent 'The 2028 Global Intelligence Crisis' report on Substack depicted a global economy destroyed by AI. The financial fiction describes the world in June 2028, where the S&P 500 is down 38%, unemployment has soared to 10.2%, and the US economy is grappling with a deflationary spiral triggered by the displacement of white-collar US workers.

The macroeconomic research firm speculates a 'human intelligence displacement spiral' or a negative feedback loop with no natural brake. In that world, AI agents swiftly replace software engineers and middle management as businesses downsize their workforce to increase margins and reinvest savings to drive AI compute growth, which further speeds up layoffs.

Citrini believes such a scenario would result in systemic financial ruin. Amid a decline in high-paying jobs, prime borrowers are also likely to default on their share of the $13 trillion residential mortgage market. Furthermore, the firm forecast volatility in private credit, predicting that private-equity-backed software-as-a-service will default on billions in debt as AI coding agents enable customers to develop internal software without paying subscription fees.

Overall, Citrini sees AI as an 'economic pandemic' generating 'Ghost GDP' in which output benefits only the owners of computing power, not trickling down through the consumer economy.

However, billionaire investor Ken Griffin's Citadel Securities debunked Citrini's dystopian outlook in a new macro strategy report using real-time economic data. The market-making giant attempted to prove that the imagined intelligence crisis was founded on a misunderstanding of macro fundamentals.

Software Jobs Continue to Rise

Although Citrini highlights the collapse of software and consulting jobs, Citadel cited Indeed job posting data showing that demand for software engineers rose 11% year-over-year in early 2026.

Citadel also used the St. Louis Fed's analysis of the Real-Time Population Survey to conclude that daily GenAI use for work remains 'unexpectedly stable' and currently 'presents little evidence of any imminent displacement risk.' Furthermore, new business formation in the US is growing fast, and AI data centre construction continues to drive hiring.

Citadel Securities said that 'despite the macroeconomic community struggling to forecast two-month-forward payroll growth with any reliable accuracy, the forward path of labour destruction can apparently be inferred with significant certainty from a hypothetical scenario posted on Substack.'

Recursive Technology Error

Citadel believes Citrini's top error is combining recursive technology with recursive economic adoption. Citrini assumes that AI integration will compound infinitely because AI can write code to improve itself, which Citadel believes is fundamentally flawed.

Griffin's firm also pointed out that Citrini overlooks the major physical limitation of energy and computing power. 'Displacing white-collar work would require orders of magnitude more compute intensity than the current level of utilisation. If the marginal cost of compute rises above the marginal cost of human labor for certain tasks, substitution will not occur, creating a natural economic boundary,' according to Citadel Securities.

Overall, capital, energy availability, and regulatory roadblocks will naturally 'brake' the 'unstoppable' feedback loop Citrini imagined, according to Citadel Securities.

Citrini also alleges that AI will destroy demand while boosting output, violating the laws of economic accounting. However, Citadel said: 'Productivity shocks are positive supply shocks: They lower marginal costs, expand potential output, and increase real income.'

For Citrini's model to play out, we must assume that labour income collapses and capital income has no velocity of spending, which is historically false. Profits from AI efficiency will be reinvested, distributed, taxed, or spent. Moreover, Citadel believes AI is likely to complement human labour rather than substitute.