The Bank for International Settlements warned in its annual economic report that the five largest hyperscalers are on track to spend more than $1 trillion on AI-related capital expenditure across 2025 and 2026, and that disappointment in returns could turn the capex boom into a protracted investment bust with knock-on effects on financial conditions.
The Basel-based institution, which advises the world's central banks, said intense competition for AI market leadership is fuelling overinvestment that resembles earlier booms tied to canals, railways, electrification and the internet. Each of those cycles ultimately delivered productivity gains, but only after periods when too much capital was committed too quickly and the unwind hit credit markets hard.
Why it matters
The scale has shifted the financing model. Where the first wave of AI buildout was funded largely from the cash piles of Silicon Valley incumbents, the current trillion-dollar cycle is leaning on corporate debt, private credit, lease financing and long-dated capacity deals. The BIS noted that AI infrastructure now reaches across corporate debt markets, private credit, data-center construction, energy contracts and supplier agreements, and that those overlapping commitments can make risks harder to see.
Compounding the strain are physical bottlenecks. Surging demand for advanced semiconductors, grid equipment and raw electricity is already pushing power prices higher, threatening to bleed into broader inflation at a moment when geopolitical conflict in the Middle East has independently stressed global supply chains. If central banks feel unable to cut rates quickly into a slowdown, the policy mix turns unusually hostile for risk assets.
Market impact
Bitcoin's first reaction would almost certainly be defensive. When liquidity tightens, investors sell what they can move fastest, and BTC sits deep in the same risk budget as equities, ETFs and high-beta assets. CryptoSlate reported Bitcoin fell below $63,000 last week after South Korea's KOSPI plunged nearly 10%, a reminder that liquidity conditions and leverage can dominate scarcity narratives for extended stretches.
The longer-arc read is more contested. Figures such as Arthur Hayes have argued that an AI bust could ultimately drive Bitcoin higher if authorities respond with renewed liquidity and investors rotate away from debt-heavy structures.
Frequently asked questions
-
What did the BIS actually say about the AI capex boom?
In its annual economic report, the BIS warned that the five largest hyperscalers are on track to spend more than $1T on AI capex across 2025 and 2026, and that disappointment in returns could turn the boom into a protracted investment bust with knock-on effects on financial conditions.
-
Why would an AI capex collapse hit Bitcoin traders first?
Bitcoin is highly liquid and held alongside equities, ETFs and other risk assets, so when portfolios are de-risked BTC is sold alongside everything else, regardless of its long-term monetary thesis.
-
What financial channels would an AI bust move through?
The BIS flagged corporate debt, private credit, lease financing, data-center construction and supplier agreements, with risks amplified by physical bottlenecks in chips, grid equipment and electricity supply.
-
Did the BIS mention crypto directly in its report?
No. The BIS did not name Bitcoin or crypto; the read-through to BTC comes from the broader mechanism of a capex-led credit shock tightening liquidity across risk assets.
-
What is the bull case for Bitcoin if an AI bust actually happens?
Figures such as Arthur Hayes have argued that an AI bust could ultimately push Bitcoin higher if authorities respond with renewed liquidity and capital rotates away from debt-heavy structures, with the caveat that the trade runs through a drawdown first.
CryptoSlate