The AI bubble: what European ETF investors should actually do (2026)

By Kestutis Balciunas·Published July 2026

The AI bubble European ETF question is on every investor’s mind: with AI-related stocks driving so much of the market’s gains, is it a bubble, and what should a European ETF investor actually do? The honest answer is that no one knows if it’s a bubble until afterward — but a globally diversified investor doesn’t need to know. This guide covers how to handle AI-concentration risk in your UCITS portfolio for 2026.

Every era has its dominant narrative — dot-com, China, FAANG, now AI. The pattern of investor behaviour around these narratives is remarkably consistent, and learning from it matters more than predicting whether AI specifically is overvalued. What follows are principles, not predictions. I cannot tell you whether AI valuations will keep climbing or correct sharply, and anyone who claims they can is guessing. What I can do is show you how to position so that either outcome leaves your plan intact.

The AI concentration reality

Whether or not AI is a “bubble,” one fact is undeniable: the major indices have become heavily concentrated in a handful of AI-exposed mega-caps. If you own a global fund like VWCE or an S&P 500 fund, you already have substantial AI exposure through the likes of Nvidia, Microsoft, and the other giants — whether you sought it or not. This concentration is the real consideration, more than the bubble debate. The AI ETFs guide covers dedicated AI exposure (which most investors don’t need on top of their index funds).

The single most useful mental shift I can offer is this: you are probably not under-exposed to AI. You are more likely over-exposed without realising it. A cap-weighted global index does not ask your permission before tilting toward whatever is winning — it simply lets the biggest companies become a bigger share of the fund. When AI is the winning theme, your “boring” global tracker quietly becomes an AI fund at the top end.

You already own the AI winners

This is the insight that dissolves most of the FOMO. A market-cap-weighted global fund holds every listed company in proportion to its size. If the AI leaders are the largest companies on earth, then they are automatically your largest holdings. You did not need to pick them, time them, or buy a thematic product — owning the market means owning the winners at exactly their market weight.

That cuts both ways, and it is worth being honest about. If AI mega-caps keep compounding, your global fund captures the upside in full proportion. If they correct, your fund takes the hit in the same proportion — but it is cushioned by the thousands of other companies you also own. A thematic AI ETF removes that cushion entirely.

A worked example: how much AI do you really hold?

Let me make this concrete with illustrative numbers (your actual figures will differ by fund and date). Suppose a broad global index fund‘s top holdings look roughly like the table below. I’m using round, illustrative weights to show the mechanism, not to quote a precise factsheet.

Add the AI-heavy lines and you might find roughly a fifth of a “diversified” global fund sitting in a handful of AI-exposed mega-caps. On a €50,000 portfolio, that’s around €10,000 already riding the theme — before you buy a single thematic ETF. The ETF overlap calculator lets you check how much of this exposure you’d be duplicating if you bolted a thematic fund on top.

The concentration risk in plain terms

Cap-weighted indices reward size, so a long winning streak in one theme mechanically increases that theme’s share of your fund. The risk is not that AI is necessarily overpriced — I’m making no claim there — but that your supposedly diversified holding has become less diversified than you assume. If a small cluster of names drives both your gains and a future drawdown, you have more single-theme risk than a glance at the fund’s name suggests.

This is precisely why “I own a global tracker, so I’m diversified” deserves a second look in 2026. You are diversified across 3,000-plus companies, yes — but weighted heavily toward the top. Understanding that distinction is the whole game.

Three ways to react to AI hype

Broadly, a European ETF investor has three sensible responses to the noise. The table below lays out the trade-offs (illustrative, not advice).

For a fuller treatment of the third option, the equal-weight vs market-cap comparison is worth reading before you act.

What history teaches

• Bubbles are only obvious in hindsight. During the dot-com era, plenty of smart people argued both sides. Calling a top in real time is nearly impossible.
• The narrative can be right and the valuations still wrong. The internet did change everything — and many dot-com stocks still went to zero. AI may be transformative and current valuations may still be excessive.
• Chasing the theme late is the classic mistake. Piling into AI-specific ETFs after the run-up is recency bias in action. See the recency bias guide.

What a European ETF investor should actually do

1. Don’t buy concentrated AI-theme ETFs chasing the trend. You already have AI exposure via your index funds; a dedicated AI ETF doubles down on concentration right after a run-up.
2. Stay globally diversified. A broad global fund spreads risk across all sectors and regions, diluting any single theme’s collapse. See the asset allocation guide.
3. Rebalance to manage concentration. If AI-driven gains have pushed your equity weight or a single holding too high, rebalancing trims it back — disciplined, not panicked. See the bull-market discipline guide.
4. Don’t try to time the pop. Selling everything to avoid a possible crash usually costs more in missed gains than it saves. Keep contributing to your plan.

To check how concentrated your portfolio has become and model rebalancing, the UCITS portfolio builder helps. If you’re weighing whether a global all-world fund already does the job, the all-world comparison hub sets out the main options.

My own experience with hype cycles

I’ve invested through three of these narratives now. In each one, the loudest temptation was to abandon a working plan and concentrate into whatever had just risen the most. The years I made the most progress were not the years I chased a theme — they were the years I kept buying my broad funds on schedule and let the index do the picking. During the last big tech surge, a reader emailed me half-convinced he needed a dedicated AI fund. We ran his global tracker through an overlap check and found he already held the same top names at roughly a fifth of his portfolio. He didn’t need a new product; he needed to see what he already owned. That conversation is the entire point of this article.

Common mistakes

• Assuming a global tracker is “safe” from AI risk. It isn’t separate from the theme — it leans into it via cap weighting.
• Buying a thematic AI ETF after the run-up. This is performance-chasing: you concentrate into exposure you already hold, right when it’s priciest.
• Selling everything to dodge a crash you’ve predicted. Market timing on a hunch usually costs more in missed gains than it saves.
• Ignoring overlap. Stacking a thematic fund onto a global fund can silently push a few names to an outsized share without you noticing.
• Treating principles as predictions. None of this tells you whether AI is over- or under-valued — it tells you how to be robust either way.

Key takeaways

• If you hold a market-cap global fund, you already own the AI winners at their market weight — no thematic ETF required.
• The real issue is concentration: a handful of mega-caps now dominate cap-weighted indices, quietly reducing your diversification.
• Buying a thematic AI ETF after a big run is performance-chasing and removes the cushion of broad diversification.
• Sensible responses are to understand your existing exposure, consider an equal-weight or ex-US tilt for balance, and resist FOMO.
• Stay diversified, rebalance with discipline, and don’t try to predict or time the bubble — robustness beats forecasting.

FAQ

Q: Is AI a stock-market bubble?
A: No one knows until afterward — bubbles are only obvious in hindsight. The narrative can be right (AI is transformative) while valuations are still excessive. A diversified investor doesn’t need to call it.

Q: Do I already own AI stocks?
A: Yes — if you hold a global or S&P 500 fund, you have substantial exposure to AI-driven mega-caps like Nvidia and Microsoft, whether you sought it or not.

Q: Should I buy an AI-theme ETF?
A: Usually not — you already have AI exposure via index funds, and a dedicated AI ETF concentrates risk further, often right after a run-up (recency bias).

Q: Should I sell to avoid an AI crash?
A: Trying to time a crash usually costs more in missed gains than it saves. Stay diversified, rebalance if concentration is high, and keep contributing to your plan.

Q: How do I manage AI concentration?
A: Stay globally diversified and rebalance if AI-driven gains push a single holding or sector too high. Diversification dilutes any single theme’s collapse.

Q: How much AI exposure does a global tracker actually give me?
A: It varies by fund and date, but a meaningful share of a cap-weighted global fund can sit in AI-exposed mega-caps. Check yours with the overlap calculator rather than guessing.

Q: Would an equal-weight ETF reduce my AI concentration?
A: It can, because equal weighting caps how large any single name becomes. The trade-off is that it may lag if mega-caps keep leading — see the equal-weight vs market-cap comparison.

Q: Does buying a thematic AI ETF on top of my global fund add diversification?
A: Generally the opposite. It usually duplicates names you already hold, increasing concentration and fees rather than spreading risk.

Q: Should I stop my regular contributions until the AI hype cools?
A: Pausing a plan to wait out a narrative is just market timing in disguise. Continuing to invest on schedule through cycles is the more reliable approach.

Q: Can you tell me when the AI bubble will burst?
A: No — and neither can anyone else with reliability. This guide is about staying robust to either outcome, not forecasting it.

The bottom line

For a European investor in 2026: whether or not AI is a bubble, you can’t know until afterward — but a globally diversified investor doesn’t need to. The real consideration is concentration: you already own substantial AI exposure through your index funds. Don’t chase AI-theme ETFs after the run-up, stay globally diversified to dilute any single theme’s collapse, rebalance to manage concentration, and don’t try to time the pop. History shows the narrative can be right while valuations are wrong — diversification handles both. The bull-market discipline guide covers managing euphoria, and the inflation hedge guide covers another macro concern.

Related reading: what to do when the world goes crazy.

Cited sources: Morningstar research, the European Central Bank, and justETF data. Educational content, not financial advice.

This guide is part of our hub on sector and thematic UCITS ETFs — browse the full cluster for related situations.

Kestutis Balciunas
European UCITS/ETF investor with 11+ years building a global dividend and index portfolio, and founder of Financial Expert Class. I write every guide here from first-hand experience investing across EU brokers and tax regimes — not theory. More about me →