Leading economists and financial analysts are cautioning that equity markets could face a downturn as enthusiasm for artificial intelligence (AI) companies reaches extreme levels. The Bank of England, the International Monetary Fund (IMF), and major investment institutions have all noted that trading in AI-focused firms has become excessively heated, increasing the chances of a wider market correction.
Why analysts worry about an AI bubble
Rapid growth in AI has pushed major technology firms — including Nvidia, Microsoft, Apple, Alphabet, and Amazon — to dominate worldwide stock indexes. Together, these five corporations now account for roughly one-fifth of the MSCI World Index, which tracks more than 1,300 companies. That concentration is about twice what occurred during the internet boom of the early 2000s. Historically, when market gains rely heavily on a small number of companies, the outcome has often been negative.
Research from investment manager GMO indicates that since 1957, the largest ten companies in the S&P 500 typically underperformed the rest of the index by about 2.4% annually. However, since 2013 those same companies have outperformed by roughly 4.9% each year. Analyst Simon Adler pointed out that similar dominance occurred during only two other periods — the dot-com era and the “Nifty Fifty” market run of the 1960s and 1970s.
Many analysts believe AI-related shares now carry valuations that exceed realistic earnings expectations. The S&P 500 trades at approximately 23 times projected earnings, compared with about 14 times for the FTSE 100, highlighting how expensive U.S. equities have become. Another valuation metric, the cyclically adjusted price-to-earnings (CAPE) ratio, has climbed above 40 — a level last seen before the technology crash two decades ago.
Adler explained that current prices are difficult to justify even under very optimistic future growth assumptions. While artificial intelligence is expected to transform many industries, investors purchasing at inflated prices risk significant losses, similar to those experienced during the 1990s internet boom.
Are today’s AI companies different?
Some specialists argue the current situation differs from earlier bubbles because leading AI businesses generate strong profits and hold substantial cash reserves, unlike speculative start-ups of the past. According to investment strategist Jason Hollands, modern technology leaders are established enterprises rather than untested ventures.
He also noted that previous speculative surges were fueled by low interest rates and heavy borrowing, whereas AI expansion has continued despite higher borrowing costs and has largely been financed through equity rather than debt. Estimates suggest spending tied to artificial intelligence could reach around $500 billion within the next year, including investment in data centers, cooling technology, and power infrastructure.
How severe could a downturn be?
While predicting the timing of a bubble’s collapse is impossible, history shows that when valuations unwind, declines can be rapid. Market crashes often produce sharp losses within weeks or months, similar to past events such as the Great Depression, the oil crisis of the 1970s, the dot-com collapse, and the 2008 financial crisis.
Since 1870, U.S. stocks have entered 19 bear markets, defined as drops of at least 20% lasting two months or longer. Following the dot-com bust, global equities declined 8% within six months and roughly 45% over three years. The United Kingdom experienced its steepest fall in 1972, when the FTSE All-Share lost about 73% during the oil crisis and recession. Recoveries from major downturns can take many years, depending on how inflated prices were beforehand.
The pandemic-era selloff was unusual — markets rebounded in roughly four months, one of the fastest recoveries in more than a century. In contrast, Japan’s Nikkei 225 required over three decades to regain its 1989 peak. After the technology bubble burst, U.S. equities did not fully recover until the mid-2010s. Generally, larger bubbles lead to longer recovery periods.
Analysts also note that current AI-driven risk is concentrated mainly in the United States, so regions such as Europe, the U.K., and Japan might experience smaller impacts, although global markets often move together. During previous crises, declines varied widely between countries, showing outcomes are never uniform.
What investors should consider
Financial advisers recommend avoiding panic and instead managing risk carefully. Diversification — investing across industries and geographic regions — can help reduce exposure to any single market. Investors should be cautious about overpriced shares and maintain a long-term perspective, as patient investing has historically produced better results than reacting to short-term volatility.