If you’ve been following the news lately, you might have noticed headlines like: “A handful of AI stocks dominate the S&P 500. If they fall, the whole market falls.”
It’s a valid concern. A key reason many investors prefer index funds is the diversification they offer. However, right now (December 2025), around 36% of the S&P 500 is concentrated in just seven companies, which is quite unusual.
So, are we witnessing an AI bubble? And if so, what should we do about it?
First off, let’s be clear: no one can definitively say whether we’re in a bubble or not. Bubbles are usually easy to identify only after they burst. What we typically observe is a new technology that could genuinely revolutionise our lives, think railways, the internet, or AI, followed by significant investments to support its development and then soaring stock prices for companies tied to that technology. Eventually, reality may set in, leading to a painful correction.
There’s no doubt that AI has real potential; major corporations are making massive investments in technologies that support it, like chips and data centres. Some AI-related companies have also shown rapid earnings growth, so it’s not merely hype. However, high enthusiasm coupled with high valuations can spell trouble for investors who jump in at the wrong time.
So, instead of trying to determine whether we’re currently in a bubble, it’s probably more useful to ask: If the leading AI companies fail to deliver on their promises, is your investment strategy still sound?
It’s essential to understand that when people talk about “the market being risky,” they often conflate two distinct concepts:
Concentration
This refers to the concern that a few stocks are dominating the index. The “Magnificent Seven” scenario suggests that if these few companies stumble, the broader index could fall too. While this is uncomfortable, history shows that concentration can change over time, and markets have still delivered good outcomes.
Valuations
This is all about the price you’re paying for a company’s future earnings. High valuations can make it tricky to expect strong returns over the long term. When you’re paying more for the same future profit, your expected returns may drop, and it’s essential to keep that in mind when investing.
A simple analogy is investing in a buy-to-let property. You might be able to buy a property for £200,000 that generates £10,000 in rent. If the same property costs £400,000 but generates the same rent, you’re paying more for the same income. You might still do well, but your return expectations should be adjusted accordingly.
Previous tech booms have shown us that even groundbreaking technologies can lead to poor investor returns if stocks are bought at inflated prices. A prime example is the dot-com era; many investors who jumped in at the height of enthusiasm didn’t see a return for a long time, despite the technological advancements.
What about my strategy?
If you have money in products such as pensions or Stocks & Shares ISAs, you’re likely investing in global equities, which typically include a large portion of US stocks. This isn’t inherently a bad thing, as the US has many exceptional companies. The risk arises when your portfolio unintentionally relies on a narrow set of assumptions:
-
That US mega-cap tech companies will continue to thrive.
-
That valuations will remain high indefinitely.
-
That the next decade will mirror the last.
These are more like assumptions than a solid investment plan.
What should I do?
A robust strategy should be designed to weather disappointments. Here’s what you can do, without needing a crystal ball:
-
Ensure Your Diversification is Genuine: True diversification isn’t just about holding 500 stocks from one country. It includes having global exposure and diverse investments across sectors and styles. A portfolio that hinges on one narrow theme is risky.
-
Set Realistic Expectations: When valuations are high, it’s wise to be prepared for modest returns rather than assuming that past performance will continue indefinitely.
