Europe needs more investment in artificial intelligence to close its productivity gap with the United States. But as AI spending soars, the European Central Bank is warning that the boom could eventually end in a correction.
In a blog published last week, five ECB economists compared today’s AI investment frenzy with previous technological revolutions, from railways and electricity to the internet. Their conclusion was not that an AI crash is imminent, but that a correction in AI-related stock valuations is likely at some point, even if today’s high prices ultimately prove justified by the technology’s potential.
“A US AI fallout would not remain a US problem.”
— ECB economists
The economists note that investors may have good reasons to pay high prices for companies whose future profits are extremely difficult to predict. But history suggests that technological revolutions tend to produce a boom and, eventually, a correction.
Europe may not escape a US sell-off
For Europe, the immediate financial risk is not so much about a home-grown AI bubble. European stock markets contain far fewer technology companies than their US counterparts, leaving them less exposed to a sharp fall in the so-called Mag7, consisting of Alphabet, Amazon, Apple, Tesla, Meta Platforms, Microsoft, and NVIDIA.
But European investors are still deeply invested in the US technology boom. Euro-area households have around €440 billion of exposure to US technology equities, much of it indirectly through mutual funds and exchange-traded funds. Pension funds and insurers are also exposed.
“A US AI fallout would not remain a US problem”, the ECB economists conclude. A correction in US technology stocks could spill into Europe through investment losses, tighter financing conditions and weaker business confidence.
You might be interested
European and US equity markets have historically been highly correlated, meaning Europe would not be insulated from a major US sell-off. As the ECB blog puts it, a correction in the Mag7 would be “a question of financial stability for the euro area, rather than just a private one”.
Investment is not the same as productivity
A similar warning came from the Bank for International Settlements (BIS) last month.
BIS economists described AI as driving a “large, increasingly debt-financed investment surge”, while cautioning that “the productivity payoff from AI, though potentially large, remains uncertain and uneven” across sectors and countries.
That distinction is particularly important for Europe. The continent needs more investment in digital technologies, computing infrastructure and AI if it wants to close its productivity gap with the United States. The answer to the AI boom is therefore hardly to step away from it. But pouring money into AI does not automatically make an economy more productive.
The ECB points out that digital investment in the euro area has grown at more than three times the pace of cumulative GDP growth over the past decade. AI adoption is also rising among European companies, despite the continent’s much smaller technology sector.
That is where Europe’s position differs from the United States. Europe is not at the centre of the current AI equity frenzy. Its markets remain dominated by more traditional industries, while its technology valuations are considerably less stretched. The ECB describes the euro area’s AI transformation as “steady if unspectacular”.
Europe cannot afford to sit this one out
None of this means Europe should step away from AI. The continent has spent much of the past decade trying to close its productivity and technology gap with the US, and AI offers one of the few technologies capable of materially changing that equation.
But South Korea offers a recent reminder of what can happen when enthusiasm for the AI boom becomes too concentrated. Its main stock index, the KOSPI, more than doubled in the first half of 2026, helped by a surge in semiconductor stocks such as Samsung Electronics and SK Hynix, before falling by around 30% from its June peak.
Retail investors had piled into the rally, often betting with borrowed money. Reuters estimates that investors lost almost $39 billion on these leveraged investments, while psychiatrists reported a sharp increase in people seeking help for stock-related distress.
Europe is in a different position. Its stock market is less dependent on technology, and the ECB says its smaller tech sector limits the risk of a home-grown crash.
But the lesson from Korea is that concentration matters. When too much of an economy’s growth, investment and investor expectations become tied to a single technology, a change in those expectations can spread well beyond the companies at the centre of the boom.
