Hussman warns: ai bubble risks bursting as debt diverges
John Hussman, the notoriously perceptive bubble-watcher and head of Hussman Investment Trust, is sounding the alarm bells again. He’s been predicting a painful correction in the stock market for years, and now, he’s pointing to a worrying divergence between corporate earnings and public and private debt as a potential catalyst for an artificial intelligence (AI) bubble implosion.
A twisted mirror image
In a weekend note, Hussman highlighted a concerning trend: a widening gap between corporate profits and debt levels. U.S. companies have seen earnings surge, fueled by the hype surrounding AI and a series of mega-deals within the tech sector. Adjusted earnings per share, after taxes, incorporating inventory valuation and capital consumption, jumped over 10% year-over-year by the end of 2025, according to data from the United States Department of Commerce. But this rosy picture is fundamentally flawed.
Hussman argues that the optimistic projections for earnings driven by AI are excessively bullish, likening the current market to a ‘Ponzi scheme.’ He notes how these markets exhibit the characteristic of paying for future cash flows that don’t exist, while simultaneously creating the illusion of attractive returns on the price investors pay. This creates a dangerous feedback loop, encouraging further investment based on unsustainable growth.
The problem, he stresses, lies in the debt – an exponential climb in household and corporate borrowing. The total federal debt recently surpassed $38 trillion, according to the United States Department of the Treasury, but the issue extends beyond the government. Households and foreign entities are also accumulating significant debt. Hussman describes the flow of corporate cash flow as a “perfectly specular image” of deficits, suggesting that corporate profits are being artificially inflated by underlying economic weaknesses.
Hussman, who accurately predicted the dot-com bubble and has long warned of a ‘third speculative bubble,’ isn’t convinced that technological advancements inherently drive sustained economic growth. He contends that the financial impact of new Technology has primarily exacerbated income inequality, necessitating more robust government support for struggling households as wealth concentrates in fewer hands. “If we allow the possibility of the United States ultimately returning to fiscal stability, then corporate profit margins will inevitably contract from their current extremes,” he speculates. ‘That’s the magic of a Ponzi scheme: everything looks good until no one questions the arrival of future cash flows.’
His firm is deliberately avoiding specific predictions regarding when the AI bubble might burst, citing the fact that historical patterns – such as peaking valuations – haven't been sufficient indicators in this instance. ‘My opinion remains that this bubble will collapse in a firestorm,’ Hussman stated resolutely. Despite growing concerns about AI investment valuations and its potential to disrupt existing business models, the sector has begun to lag behind other areas like energy and materials. The market’s recent vertical ascent – peaking just 72 hours before the next crucial trading window – underscores the precariousness of the current situation.