The AI Boom Looks Like a Bubble in Some Ways and Nothing Like One in Others

The five largest companies made up about 30% of the S&P 500 in late 2025, the highest concentration in 50 years, and the index traded at 23 times forward earnings. Unlike the dot-com era, today's leaders are highly profitable. The case on each side.

A handful of technology companies now carry a larger share of the US stock market than at any point in half a century. Photo: Dietmar Rabich, via Wikimedia Commons (CC BY-SA 4.0)

By the UISC BD Editorial Desk · United Information Service Center · Published 13 September 2026 · 7-minute read

This article sets out publicly reported data and arguments. It is general information, not financial or investment advice, and it does not predict market direction.

"Is this a bubble?" is the question hanging over global markets. The honest answer is that the evidence points both ways, and anyone who sounds certain is guessing.

The Case That It Looks Stretched

Concentration. By late 2025, the five largest companies made up about 30 percent of the S&P 500 — the greatest concentration in 50 years. The ten largest accounted for roughly 35 to 41 percent, depending on the date, and the so-called Magnificent Seven about 35 percent.

Valuation. In early 2026, the S&P 500 traded at around 23 times forward earnings, the most stretched level since the dot-com era.

Spending ahead of revenue. Estimates put AI infrastructure spending at roughly $400 billion, against about $100 billion in AI revenue. Data centres, chips and power have to earn a return eventually — and their electricity demands are substantial.

The Case That It Is Different

Profits are real. At the peak of the dot-com bubble, only around 14 percent of the companies involved were profitable. Today's leaders are among the most profitable businesses in history.

Nvidia as the example. The chipmaker at the centre of the boom reported fiscal 2026 revenue of $215.9 billion, up 65 percent, with a net margin of about 53 percent. That is not a company selling a promise.

Demand is visible. Businesses and consumers are actually using AI tools at scale — the question is how much they will pay over time, not whether the technology exists.

Both Can Be True

A technology can be transformative and its shares can still be overpriced. The internet changed the world, and many internet stocks still collapsed in 2000. Railways, electricity and cars all went through investment booms that overshot before the technology delivered.

The useful question is not "Is AI real?" It is "Are today's prices already assuming more growth than will arrive?" Markets will answer that over years, not weeks.

What It Means for Ordinary Investors

Index funds are more concentrated than they look. Someone holding a broad US index fund has a large share of their money in a handful of technology companies, whether they chose that or not.

Interest rates matter. High valuations are especially sensitive to rate expectations — relevant ahead of the Federal Reserve's 16 September decision, and explained further in how rate changes reach your money.

Hype attracts scams. Every boom produces fake "AI investment" schemes riding the headlines — see what regulators say about AI trading bots.

Diversification is the traditional answer to uncertainty. No one knows how this ends, which is exactly why spreading risk exists.

Sources

"AI bubble vs dot-com comparison," IntuitionLabs — intuitionlabs.ai
"AI mania is fueling bubble-like concentration in the S&P 500," Yahoo Finance — finance.yahoo.com
"2026 market outlook: if we're in an AI bubble, it's a profitable one," Seeking Alpha — seekingalpha.com
Nvidia earnings and the AI trade, Yahoo Finance — finance.yahoo.com

ai bubble 2026 (1)

Is the AI Stock Boom a Bubble? The Numbers on Both Sides