AI bust could hit banking and credit markets, BIS warns

The Bank for International Settlements warns that an artificial-intelligence bust could ripple from economic growth into banking and credit markets, as more than $1 trillion in hyperscaler spending outpaces earnings and forces some firms to issue debt across opaque financing chains. In its 2026 annual report, the Basel-based institution said disappointment in AI returns could trigger a sudden financing pullback, supply-chain defaults, and risk repricing with wider macroeconomic fallout.

The warning landed Sunday, on the eve of the European Central Bank's annual symposium in Sintra, where global policymakers were set to scrutinize financial stability risks. For households watching portfolios and policymakers watching systemic risk, the message is blunt: the AI capex boom is no longer just a tech story. It is a banking and credit story too.

Key Takeaways

• The BIS says the five largest hyperscalers are on track to spend more than $1 trillion on AI-related capital expenditure across 2025 and 2026, outpacing earnings and free cash flow.
• Officials warn that circular financing, private credit, and debt-funded data center buildouts could amplify losses if AI returns disappoint.
• An equity correction could hit consumption harder than in past tech busts because household stock exposure has risen since the dot-com era.
• The BIS compared today's AI frenzy to canal, railway, and dot-com manias that ended in recessions, but stopped short of prescribing immediate rate hikes.
• Policymakers urged stronger oversight beyond traditional banking, including non-bank lenders financing AI infrastructure.

What did the BIS say about the $1 trillion AI buildout?

In its Annual Economic Report 2026, the Bank for International Settlements — often called the central bank for central banks — placed AI sustainability among the most urgent pressure points facing the global economy. The institution cited risks of persistent inflation, weakening fiscal positions, growing financial vulnerabilities, and whether AI-related investment can deliver returns commensurate with today's spending pace.

According to reporting from Bloomberg, the BIS highlighted AI-led risks prominently in a report that arrived as markets were already debating whether big tech valuations had run ahead of fundamentals. Fortune noted that BIS economists drew explicit parallels to historical technological manias: the canal boom of the 1830s, the British railway bubble of the 1840s, and the dot-com crash of 2000.

Each episode, the BIS argued, began with a genuine breakthrough that attracted more capital than commercial returns could ultimately justify. Each ended with an investment reversal that contributed to economy-wide recessions. The institution wrote that the scale and pace of the current AI investment boom, accompanied by expectations of large productivity payoffs, bears resemblance to those precedents.

How could an AI bust hit banking and credit markets?

The transmission channel runs through leverage, interconnected borrowers, and compressed risk premia. The five largest hyperscalers are set to spend over $1 trillion on AI-related capex from 2025 through 2026, according to the BIS. Those commitments are outpacing earnings and free cash flow, pushing some firms to issue debt to close the gap.

That shift matters for banking and credit conditions well beyond Silicon Valley. The BIS warned that if hyperscalers slow or halt aggressive capex deployment, borrowers across the supply chain — infrastructure contractors, chipmakers, AI labs, and private credit lenders — could struggle to replace lost revenue and service debt. Engineering and construction firms at the end of the chain are particularly vulnerable, with comparatively weak balance sheets and limited cushion against a sudden reversal.

Officials also flagged circular financing arrangements linking hyperscalers, chip makers, and AI labs through equity stakes and multi-year purchase commitments. Data center construction is increasingly outsourced to third parties that lease facilities back on long-dated contracts with embedded exit clauses. The BIS said such deals are often poorly disclosed, with risks that the same asset may be pledged multiple times.

Bloomberg reported that repricing of risk — whether triggered by higher interest rates or an AI bust — has the potential to be similarly disruptive in credit markets to the 2008 global financial crisis. The Telegraph similarly reported that excessive, debt-fuelled spending on AI data centers and opaque transactions raised the risk of a financial meltdown reminiscent of the global credit crunch nearly two decades ago.

The BIS also noted stress already visible in parts of private credit. Direct lending funds catering to retail investors have faced mounting redemption requests, forcing some to liquidate assets despite having no contractual obligation to do so. Banks' growing and opaque exposure to private credit funds, compounded by overlapping ties through insurance company balance sheets, could spread turmoil beyond the non-bank perimeter.

Why are household portfolios part of the risk picture?

Equity markets are another amplification channel. The BIS observed that a major equity-market correction could have larger macroeconomic consequences today than in the past because US households have increased their exposure to stocks since the dot-com bust. Rich valuations, especially among AI-linked firms, mean sentiment could reverse quickly if earnings growth disappoints.

Fortune reported that traders were already navigating volatility in late June 2026, with major US indexes under pressure amid questions about whether AI leaders could sustain spending and deliver returns. The BIS said disappointment in returns could trigger a sudden pullback in financing and turn the capex boom into a protracted investment bust, with knock-on effects on financial conditions.

Inflation adds a policy dilemma. AI's demand for electricity, advanced semiconductors, and grid equipment is generating input cost pressure at a moment when energy supply shocks have pushed inflation above target in major economies. Yet the BIS declined to recommend rate hikes as a direct response, citing extreme uncertainty. BIS General Manager Pablo Hernández de Cos said attempting to be prescriptive about monetary policy would be unwise given the scale of unknowns.

What should investors and savers watch next?

The BIS did not forecast an imminent crash. It framed the risk as conditional: if AI payoffs fall short, financing structures that look manageable during exuberance can unwind abruptly. That makes balance-sheet quality, debt maturity, and exposure to a narrow set of hyperscaler customers more important for anyone building long-term wealth.

For readers focused on passive income and portfolio resilience, the lesson is diversification across asset classes and vigilance about concentration in AI-linked equities and credit. If you are rethinking how macro shocks affect income strategies, see more coverage in our Wealth Hacks & Passive Income section.

Central bankers are not calling the top on AI. They are warning that the same boom supporting growth today could become a source of financial instability tomorrow if banking channels, private credit markets, and household wealth all move in the wrong direction at once. The next data points to watch are hyperscaler capex guidance, credit spread moves among AI borrowers, and whether inflation forces tighter policy into an already stretched financing cycle.