The Stock Market Has Not Been This Expensive Since the Dot-com Bubble's Peak. History Says to Prepare for What Might Come Next.

Every day, the financial media bombards the world with debates about whether the artificial intelligence ( AI ) bull market has turned into a bubble. Pundits will go on TV and loudly support one side or the other in this argument, often with little fundamental analysis to back them up. This can leave viewers with few ways to assess the stock market's condition outside of vibes.

But how exactly can you quantitatively define when the stock market is overvalued? The best metric to use might be the Shiller Cyclically Adjusted Price-to-Earnings ratio , otherwise known as the Shiller CAPE ratio. And that metric just hit its most expensive level since the dot-com bubble's peak in late 1999 and early 2000.

Missed Nvidia in 2009? This Rare Signal Is Flashing Again. In 2009, a "Double Down" signal flashed for a little-known chipmaker called Nvidia. For the first time in years, that same "Total Conviction" signal is flashing for a company 1/100th the size of Nvidia. Continue »

Here's what that could mean for the AI bull market.

Unlike the traditional P/E ratio, which simply takes a company's current stock price and divides it by trailing earnings per share (EPS), the CAPE ratio takes a longer-term view to measure earnings and valuation through economic cycles.

Specifically, the CAPE ratio can be applied to something like the S&P 500 index. The numerator of the ratio will remain the same: the combined share price of all the stocks in the index, weighted by market capitalization. But the denominator will be the average EPS -- adjusted for inflation -- over the last 10 years, rather than just the trailing 12 months.

The primary reason investors focus on the market's CAPE ratio instead of its simple P/E ratio is that it's designed to smooth out earnings over a business cycle, where one year may run hot (for example, 2026) and others may run cold (for example, 2020 during the pandemic lockdowns).

The S&P 500 index now trades at a CAPE ratio of over 41, its highest level in history outside of the period at the end of the dot-com bubble in late 1999 and early 2000. This should give even an ultra-bullish investor pause when weighing the question of whether AI stocks are in a bubble.

In 2026, S&P 500 earnings have soared due to rising spending on semiconductors, memory, and AI software, and the rising valuations of start-ups like OpenAI and Anthropic. When big tech companies invest in OpenAI and OpenAI's value climbs in a quarter, those investment gains register as earnings for that quarter. These are one-time benefits that are unrelated to their underlying businesses, though, and they are a key reason analysts estimate that the S&P 500's overall earnings grew by 52% year-over-year in Q2. However, this type of growth is not sustainable and inflates the trailing P/E figure.