- In early November tech stock markets fell on doubts about AI valuations.
- The worry was that profits would not justify prices and that data-centre spending would not turn into revenue in time.
- Bubbles are better recognised in hindsight: what can be done is to diversify, avoid borrowing to invest and keep liquidity.
In early November 2025, tech stock markets fell on doubts about the valuations of artificial intelligence companies and about the huge spending on data centres. After months of almost continuous gains, the market was beginning to ask an uncomfortable question: whether prices had gone too far and whether investments would turn into revenue in time.
These doubts were not new, but they gained weight as the announced figures grew larger. The conversation about a possible bubble took hold in the media and in portfolios, and with it the temptation to try to time entry or exit. It is best approached with more calm than anxiety, and by understanding what can and cannot be known. Nobody knows for certain when a high valuation becomes a problem, and anyone who claims otherwise deserves some scepticism.
What the market doubted
The falls reflected two concerns: that the valuations of artificial intelligence companies had soared beyond what their profits justified, and that the huge spending on data centres would not turn into revenue in time.
Both concerns point to the same risk: that expectations have run ahead of results. When that happens, a disappointment, a delay or a mere downward revision of forecasts is enough for the market to reconsider prices. That does not mean the technology has no future, but that its present price may reflect excessive optimism. It may also simply reflect growth that has not yet materialised but will come.
What a bubble is
Bubbles are better recognised in hindsight. There is a bubble when prices stray far from what fundamentals justify, but at the time it is hard to know whether a rise reflects a real structural change or an excess of optimism, and so it is prudent not to bet everything on a single reading.
History offers examples of technologies that changed the world and, at the same time, of periods in which their share prices were unsustainable. Both can be true at once: an innovation can be real and its valuation excessive. It is therefore wise to separate the question of whether a technology has a future from that of whether a specific price reflects it reasonably. The first is a question of technology and business; the second, of price and patience.
What can be done
What can be done in good time is to diversify, avoid borrowing in order to invest, and keep liquidity to take advantage of corrections. These are simple principles, but hard to follow when the environment is euphoric, which is why it is wise to set them in advance, with an adviser, instead of improvising them under pressure.
A practical guide is to ask what part of one’s wealth could withstand a sharp fall without altering one’s life plans, what part would be needed in the next few years and what part can remain invested for the long term. With those answers, the decision about how much risk to take stops depending on headlines and starts to depend on one’s own goals. That brings more peace of mind, which at such moments has a value of its own. None of this is a recommendation about any security.
Discipline is defined calmly, before the market puts it to the test.
This article is general information. It is not legal, tax or investment advice, and it does not take anyone’s particular circumstances into account.
