The Bank for International Settlements (BIS), often described as the central bank of central banks, warned this week that the current surge in investments tied to artificial intelligence could result in a prolonged downturn. In its commentary the BIS compared the situation to large historical investment waves — canals in the 1830s, railways in the 1840s and the dot‑com boom of the 1990s — episodes that brought massive capital inflows but produced returns that could not fully justify the spending.
Who and what did the BIS highlight?
The BIS named as a key factor that five hyperscalers are expected to spend more than $1 trillion in total through 2026. The bank treats that scale of investment as a source of systemic vulnerability if returns or financing conditions deteriorate.
What are the macroeconomic risks?
According to the BIS, much of the AI investment boom has been fuelled by cheap credit. That financing structure means that a weak quarter in returns or a tightening of credit conditions could rapidly freeze investment activity, triggering liquidity stress and asset price corrections. Unlike some past bursts, BIS noted, this episode could affect a broader set of economic actors because household equity holdings are at historically high levels.
That higher household exposure increases the chance that any market correction will spill over into consumption and the real economy, rather than remaining confined to institutional investors or technology firms.
Why this matters
The BIS’s warning carries weight because of the institution’s role in advising and monitoring central banks. When the BIS signals a risk of a bubble and prolonged bust, policymakers and market participants often take notice. By invoking historical parallels, the bank underscored that major technological breakthroughs do not by themselves ensure that the large volumes of capital deployed will be profitable.
Brief conclusion
The BIS message is cautious: the rapid, large‑scale flow of capital into AI and related infrastructure could produce significant economic knock‑on effects if returns disappoint or financing tightens. Given the scale (more than $1 trillion by five hyperscalers through 2026) and high household equity exposure, the bank recommends that both regulators and market actors account for the risk of a substantial correction.
