A familiar boom-bust warning indicator that preceded both the dot-com crash and the 2008 housing crisis is now worth monitoring within the artificial intelligence sector.
Referred to as a “funding gap,” this market signal is examined across two previous historical downturns in a report by the Wall Street Journal. Columnist Jonathan Weil contends that the ongoing AI trade relies heavily on this exact continuous injection of capital.
The Boom-Bust Signal in 2000 and 2007
While Weil points out various other challenges threatening the AI market—including potential disruptions from smaller competitors and power shortages at data centers—he emphasizes that a narrowing capital window poses the most pressing danger to observe.
As the dot-com bubble began to deflate in 2000, terms like “funding gap” and “burn rate” became widespread. For instance, Pets.com persistently secured new funding even while marketing its goods at deep losses, until financing dried up completely and the firm collapsed just months afterward.
A comparable dynamic unfolded in November 2007, when Freddie Mac reduced its dividend and issued $6 billion of preferred stock in an effort to reinforce its balance sheet following escalating mortgage-related losses.
Despite these measures, Freddie Mac failed to secure the common equity that investors demanded. That shortfall contributed directly to the financial crisis that culminated in the enterprise being placed into government conservatorship the subsequent year.
Is the AI Trade Showing the Same Cracks?
Based on the Journal’s findings, leading firms like OpenAI and Anthropic are currently burning through cash without generating public profits. Furthermore, OpenAI’s capacity to keep drawing in funds relies in part on ongoing support from SoftBank and Nvidia (NVDA).
Meanwhile, Anthropic’s tentative timeline for an initial public offering has already moved from October to November. Additionally, Holtec Nuclear—a key provider for AI data centers—put its own stock listing on hold this month.
Even so, the Journal points out that these postponements currently reflect preliminary caution rather than definitive evidence that the financing window has shut. At the same time, Oracle’s (ORCL) borrowing-financed investments in data centers rely on OpenAI’s continued ability to pay for computing power, thereby linking the broader AI market directly to a small group of private laboratories.
The Journal observes that market booms can unwind gradually or come to an abrupt halt. Because both previous crashes became apparent the moment the concept of a funding gap re-emerged in discussions, the market may ultimately only identify whether that same phrase applies to AI through hindsight.