At a Glance
- Are we in an AI bubble? No one can know in advance—and long-term investors don’t need to. Predicting bubbles is far less important than building portfolios that can endure them.
- Today’s AI boom is different from past bubbles, but market concentration is real. A small group of companies now drives a large share of index returns and capital spending.
- Diversification remains a practical, real-time tool, helping manage concentration risk while allowing portfolios to adapt as market leadership inevitably changes.
Are We in an AI Bubble?
It has been nearly three years since the arrival of OpenAI’s ChatGPT-3.5, marking generative AI’s watershed moment. Suddenly, algorithms could produce text, computer code and images comparable to human output—and an AI investment boom was underway. Fast-forward to today, when the investment news is filled with comparisons to the dot-com bubble of the late 1990s and questions about whether the boom may soon turn to bust.
As in the late 1990s, a transformational technology has sparked enormous enthusiasm and aggressive capital expenditures (capex). For 2025, big tech spending clocks in at an estimated $400 billion. Building the infrastructure required to support this technology is expected to cost $3 trillion through 2028. Some worry that the technology ultimately will not provide enough value to justify the investment. OpenAI, for instance, is planning a $500 billion data center project, even though the company will generate only $13 billion in revenue in 2025.
Yet there are important differences between 2025 and 2000. Unlike the speculative companies of the dot-com era, today’s biggest public technology firms are highly profitable and funding capex out of substantial cash flows. And while valuations are elevated, they’re not at the extremes seen in 2000. The S&P 500 Information Technology Index recently traded around 30 times forward earnings, well below the dot-com era peak of 55.
Should You Worry About a Bubble?
Reasonable arguments exist on both sides of the bubble debate. But long-term investors don’t need to pick a side. Correctly identifying a bubble is extraordinarily difficult—and it’s unnecessary.
Your job is not to figure out whether a particular market is moving too far, too fast. It’s to invest in a way that gives you the best chance to reach your long-term goals. The key to that task is to build and maintain a portfolio that can keep you on track toward those objectives across many different market environments, including both booms and busts. That means diversifying across asset classes, sectors, company sizes and geographies.
Diversification to Balance Risk and Potential Reward
Many investors assume their stock holdings are well diversified if they track the S&P 500. However, the so-called “Magnificent Seven”—seven of the index’s largest technology companies—now account for roughly 35% of the index. Those same companies account for about 30% of all capex in the S&P 500, a large share of which is AI-related. In other words, mirroring the S&P 500 means you’re betting a significant chunk of your future on AI-driven growth. If the boom hits a speed bump, you might be over-exposed to the downside.
That doesn’t mean avoiding innovation or transformative technologies. It means being deliberate about how much of a portfolio’s future is tied to a single narrative. Diversification can help limit the risk of this type of concentration.
In practice, diversification is not about owning everything equally. It’s about continually assessing where capital is becoming crowded, where expectations are extreme, and where future returns may be more resilient. The dot-com bust offers a useful case study of the ways small-cap and international equity allocations can help reduce the impact when large growth stocks decline.
The dot-com bubble burst in March 2000, sending large growth stocks into a freefall. Over the five years through March 2005, the Russell Top 200—the market’s 200 largest stocks by market capitalization—lost more than 25%.[1] Meanwhile, the Russell 2000 index of small caps did almost exactly the opposite, gaining about 23%[2] over the same time period. International stocks also outperformed, beating U.S. stocks between 2000 and the 2008 financial crisis. The upshot: Investors with diversified portfolios had a very different, less turbulent experience than investors who concentrated on the stocks that dominated the indexes at the end of the 1990s. Spreading their investments around may have supported their account balances during the first half of the 2000s, possibly leaving them with more assets to benefit from subsequent gains.
Long-Term Investors Don’t Need to Predict Bubbles to Manage Risk Intelligently
We’re not predicting that history will repeat itself. No one knows what the future holds. The point is that market leadership can change, sometimes abruptly, and a diversified portfolio is designed to adapt to those changes. With a diversified portfolio that’s built around your goals, you don’t have to predict when or why such shifts will occur.
Bubbles are clear only in hindsight. Diversification, on the other hand, works in real time. And it remains one of the most effective tools you have to navigate uncertainty. As the new year begins, periods like this are often a useful time to revisit portfolio structure, concentration, and assumptions—not to make bold bets, but to ensure your capital is positioned thoughtfully for whatever comes next.
[1] Cumulative return calculated from -5.73% annualized return for the five years through March 2005.
[2] Cumulative return calculated from 4.30% annualized return for the five years through March 2005.
Past performance does not guarantee future results. All investments include risk and have the potential for loss as well as gain.
Data sources for returns and standard statistical data are provided by the sources referenced and are based on data obtained from recognized statistical services or other sources we believe to be reliable. However, some or all information has not been verified prior to the analysis, and we do not make any representations as to its accuracy or completeness. Any analysis nonfactual in nature constitutes only current opinions, which are subject to change. Benchmarks or indices are included for information purposes only to reflect the current market environment; no index is a directly tradable investment. There may be instances when consultant opinions regarding any fundamental or quantitative analysis do not agree.
The commentary contained herein has been compiled by W. Reid Culp, III from sources provided by TAGStone Capital, as well as commentary provided by Mr. Culp, personally, and information independently obtained by Mr. Culp. The pronoun “we,” as used herein, references collectively the sources noted above.
TAGStone Capital, Inc. provides this update to convey general information about market conditions and not for the purpose of providing investment advice. Investment in any of the companies or sectors mentioned herein may not be appropriate for you. You should consult your advisor from TAGStone or others for investment advice regarding your own situation.
