- On 27 January, the Chinese model DeepSeek, trained with far fewer resources than usual, shook the artificial intelligence sector.
- Nvidia shares fell 17% in a day, the largest loss of value for a company in a single session.
- The lesson is general: concentrating wealth in a single thesis, however solid it looks, is a risk worth reviewing with an adviser.
On 27 January 2025, the appearance of the Chinese artificial intelligence model DeepSeek, trained with far fewer resources than usual, shook the markets. Nvidia shares fell 17% in a single day, the largest loss of value for a company in one session. The episode was a reminder that in markets certainties can crumble very quickly.
Until then, a good part of technology investment had been built on a widespread idea: that progress in artificial intelligence required enormous quantities of chips and computing capacity. The news forced a review of that premise and showed how far the expectations of an entire sector can depend on a single narrative. When many investors share the same reading of a sector, the surprise spreads quickly and affects not just one company but all those that depend on the same assumption.
What scared the market
The dominant idea was that artificial intelligence required gigantic investment in chips and data centres. A model trained with far fewer resources cast doubt on that premise and, with it, on the revenue expectations of the companies that supply that equipment. If less hardware is needed, future demand for those who make it may be lower than forecast.
A share price sums up what the market expects from a company over years, not only what it earns today. That is why news that changes those expectations can move the price at a stroke, without the company having sold less or made any mistake. The adjustment, when it comes, tends to be sudden rather than gradual.
Advantages that erode
The market discovered that technological advantages can erode quickly. A lead that seems unassailable can be challenged by a competitor with a different approach, different costs or a different way of solving the same problem. In fast-changing sectors, today’s advantage is no guarantee for tomorrow.
This also holds outside technology. Any company with a dominant position, a patent, a brand or a process that is hard to copy should ask itself what would happen if someone did the same with half the resources. Looking at that question honestly is a good management tool, whether one is an investor, a director or a business owner. The answer need not be pessimistic: sometimes it confirms that the advantage is solid, and other times it points to what should be strengthened before a competitor does so.
The lesson for a portfolio
Beyond this case, the lesson is general: over-concentrating a portfolio in a single thesis, however solid it seems, is always a risk. It is worth reviewing from time to time how much wealth depends on a single sector, country or idea, and talking to an adviser about how to spread that risk according to each person’s goals and horizon.
This does not mean avoiding risk, which is part of investing, but knowing it and choosing it consciously. A diversified portfolio withstands surprises better, and a periodic review helps to detect whether a position has grown until it weighs more than intended. It also helps to set down in writing, in advance, the maximum proportion accepted for each theme, so that the decision does not depend on the mood of the moment. None of the above is a recommendation about any particular security.
The most widely accepted theses are the ones least questioned, until the day someone questions them.
This article is general information. It is not legal, tax or investment advice, and it does not take anyone’s particular circumstances into account.
