AI Giants Hide $1.65 Trillion in Debt, Echoing Enron Tactics

The $1.65 Trillion Blind Spot in AI's Spending Spree

The artificial intelligence arms race is fueling an unprecedented construction boom, with tech giants pouring hundreds of billions into data centers. However, a new investigation reveals a staggering financial reality: the true cost is largely hidden from plain sight. According to Nikkei Asia, five major US tech companies—Alphabet, Amazon, Meta, Microsoft, and Oracle—have amassed an estimated $1.65 trillion in off-balance-sheet debt, a figure that eclipses their combined $1.35 trillion in officially reported liabilities.

This hidden debt, which has surged eightfold in just four years, stems from opaque financing arrangements for AI infrastructure. These include long-term leases for data centers and massive supply contracts for GPUs. The practice allows companies to keep massive financial obligations off their primary balance sheets, painting a potentially misleading picture of their fiscal health to investors and analysts.

How the Debt is Hidden: A Return to Enron-Era Tactics

The mechanism for this financial obfuscation is strikingly familiar. Companies are using special purpose vehicles (SPVs) and legally distinct subsidiaries to finance their AI buildout. This allows them to classify what are essentially debts as operating leases or purchase commitments, which are disclosed only in the footnotes of SEC filings rather than as headline liabilities. This accounting treatment has drawn direct comparisons to the notorious collapse of Enron in 2001.

“The accounting treatment itself is in fashion,” technical accounting consultant Tom Selling told Bloomberg. “But what if one of these companies was a house of cards and was propping itself up with this accounting treatment? To me, that’s the risk.” The comparison is potent, as Enron famously used similar shell companies to hide billions in debt before its spectacular implosion, a cautionary tale that now echoes in the AI sector.

Meta and Oracle Lead the Hidden Debt Surge

The scale of the hidden obligations varies dramatically among the companies. Meta Platforms stands out with an estimated $420 billion in off-balance-sheet debt, nearly triple its transparent debt. This figure alone underscores the immense financial commitments required to compete in the AI race. Oracle's situation is even more extreme, with its off-balance-sheet commitments expanding 30-fold in just four years, highlighting the company's aggressive pivot toward cloud and AI infrastructure.

For investors, the problem is one of visibility. Relying solely on standard debt metrics like net debt or leverage ratios provides an incomplete and potentially dangerous view. The hidden obligations, while disclosed in regulatory filings, are buried in complex footnotes that many market participants overlook. This opacity makes it difficult to accurately assess the true risk exposure of these tech behemoths.

continue reading below...

The AI Bubble and the Funding Gap

The revelation of this hidden debt comes amid growing fears of an AI bubble. Experts have long warned of a widening gulf between sky-high company valuations and the comparatively meager profits generated by AI services. The latest data from Morgan Stanley and Moody’s further highlights the issue, with both agencies flagging the risks associated with the opaque financing structures. The market has “begun to show concern,” according to Nikkei.

However, the narrative is not one of simple financial distress. The $1.65 trillion figure is often misread. As Forbes clarifies, this is more accurately described as a funding gap—the difference between the $2.9 trillion in global data-center investment needed through 2028 and the $1.4 trillion that can be funded by Big Tech's cash flow. The gap is being filled through a mix of bonds, private credit, and asset-backed securities. Morgan Stanley forecasts about $570 billion in global AI-related debt issuance in 2026 alone.

The Real Risk: Who Holds the Bag?

The key question is not whether the debt exists, but who will bear the losses if the AI boom falters. If enterprise AI spending disappoints and data centers fail to generate projected returns, the hidden lease commitments could trigger massive impairment charges. These write-downs would directly hit shareholders, but the pain would also spread to lenders, insurers, and private-credit investors who financed the construction.

Bond investors are already showing signs of pushback. Demand for AI-related debt is softening as the market absorbs a record wave of issuance. This creates a precarious situation: the very financing that enables the AI buildout could become a source of systemic risk. As 24/7 Wall St. notes, “investors who rely only on headline debt figures are missing a large portion of the AI spending story.” The quiet question is who ends up holding the risk when the music stops.

What to Watch Next

Four of the five companies in the Nikkei study are set to report their second-quarter earnings in the coming days. Investors and analysts will be scrutinizing financial statements not just for profit figures, but for the footnotes that reveal the true scale of AI-related commitments. The pressure is on for these tech giants to justify their massive spending and demonstrate a clear path to profitability.

The AI industry is at a critical inflection point. The technology holds immense promise, but the financial engineering behind its infrastructure buildout carries echoes of past market excesses. The coming weeks will be crucial in determining whether this is a calculated investment in the future or a house of cards waiting to fall.