BIS Warns $1 Trillion AI Spending Bubble and Shadow Banking Ties Could Trigger 2008-Style Global Financial Crisis

1. Introduction: The BIS Warning

The Bank for International Settlements (BIS)—frequently referred to as the "bank for central banks"—has released a major warning in its Annual Economic Report. The institution warns that the current frenzy in artificial intelligence capital expenditure, combined with unregulated "shadow banking" debt pipelines and circular financing, could trigger a systemic credit crunch similar to the 2008 global financial crisis.

As major tech conglomerates borrow billions of dollars to build infrastructure for models whose commercial viability remains unproven, any sudden correction in AI expectations could send shockwaves through the global credit and banking systems.

2. The $1 Trillion CapEx Binge of Tech Hyperscalers

According to the BIS report, the world's five largest tech "hyperscalers" (Alphabet, Amazon, Meta, Microsoft, and Oracle) are on track to spend a combined **$1 trillion** on AI-related capital expenditure (CapEx) between 2025 and 2026. This spending is increasingly outpacing their actual free cash flow, prompting them to turn to the credit markets.

The rapid deployment of AI data centers requires massive upfront investments in high-end processors, cooling systems, and power grid connections. While these capital investments are driven by intense competition, the cash flows generated by actual AI product subscriptions (like ChatGPT Plus or Copilot Pro) are still far too low to justify the scale of the hardware spending. For context on this infrastructure race, read our analysis on Google Gemini compute limits forcing Meta onto internal models.

3. The Peril of "Circular Financing"

A key vulnerability identified by the BIS is the growing prevalence of "circular financing." In these arrangements, major chipmakers and cloud giants invest equity capital in startup AI laboratories. In return, the AI labs commit to spending that investment capital on purchasing those same investors' chips and cloud compute power.

This creates a closed financial loop where capital is recycled back to the original hardware providers as artificial revenue. This practice hides the true cost of AI development and exaggerates commercial demand, leading to inflated evaluations and severe asset bubbles.

4. Shadow Banking: The Loose Funding Channels

Unlike previous technological booms, a large portion of the current AI infrastructure buildout is being funded outside of the traditional regulated banking system. Hedge funds, private credit vehicles, and shadow banks are providing billions in loosely regulated debt to underwrite data center projects.

Because these shadow banking networks operate with minimal regulatory oversight, a default in the tech sector could transmit and amplify risk across the global financial system much faster than normal banks, creating hidden leverage vulnerabilities.

5. Macroeconomic Risks: A Potential 2008-Style Bust

The BIS General Manager, Pablo Hernández de Cos, warned that the current AI boom shares striking similarities to historical financial manias, such as the railway mania of the 19th century and the dotcom bubble of the late 1990s. If the returns on these mega-investments continue to disappoint, or if tech hyperscalers begin to scale back their capital deployment, the market could experience a sharp contraction.

Because tech debt is heavily integrated into global bond markets, a sudden repricing of tech equities and corporate debt could lead to a broader credit squeeze, threatening general economic growth. Central banks are urged to implement strict oversight on non-bank financial intermediaries before any correction occurs.

6. Frequently Asked Questions (FAQ)