Some market signs are now truly concerning. The data I’ve been seeing over the past few weeks is sending a clear message all together: the S&P 500 has now reached an extremely risky valuation level.

First, let’s talk about the CAPE ratio. This is an indicator created by Robert Shiller based on average earnings over the past ten years. In February, it reached 39.2. This level has only been seen historically in just two periods: at the peak of the dot-com bubble in 2000 and at the end of the 1920s. When this ratio is this high, future returns are generally very weak—currently only about 2% annually.

The second warning sign is the Warren Buffett indicator. This is the ratio of market capitalization to GDP. Buffett himself said it is the best single measure for valuation. Now it has exceeded 230%—the highest in history. When he proposed it, he considered 75%–90% to be reasonable. A value above 120% means the market is overvalued. Now we are at double that level.

The third problem is market concentration. The top ten stocks now account for more than 35% of the S&P 500. The Magnificent Seven alone make up more than 30%. This means that when a few of these stocks fall, the whole index falls with them. We have already seen this—this year, Microsoft, Amazon, and Nvidia have declined by 20%, 9%, and 6%, respectively, which has caused the index to drop by more than 4%.

There are fierce disagreements on Wall Street about this. One camp says it’s a big bubble. GMO’s Jeremy Grantham has clearly stated that AI investment is actually not turning profitable. OpenAI itself is predicting that it will lose 170 billion dollars in 2026. Some investment management firms, such as Kathmere Capital Management, are staying alert about these valuation concerns.

On the other hand, the bullish camp says that profit growth justifies this valuation. Analysts expect profits to grow by 17.6% in 2026. Morgan Stanley noted that most bull markets last 5–7 years, and in the fourth year, there have always been positive returns in history.

But the big question here is: will this AI capital expenditure actually be converted into profits? And will geopolitical crises—( Iran, oil prices )—turn into an economic recession? If both happen, historically the S&P 500 has fallen by as much as 32% on average. If profit growth continues, the correction may be limited. Now the market is caught between these two opposing messages.