As we approach the end of 2025, Wall Street is caught between two powerful influences: increasing doubts about the AI-driven purchases that have fueled this year's rally and the historical seasonal effects that have lifted the markets in December for nearly a century.
This tension has divided investors: Should they chase the rally or prepare for a possible correction?
Crowded trades do not make easy money.
The 'Santa Claus rally' encompasses the last five trading days of December and the first two trading days of January. It has provided nearly a 79% return since 1929; the average gain stands at 1.6%. In the last eight years, there has only been a loss once.
However, some skeptics now believe that the seasonal effect has become overly clichéd. One investor wrote on X: 'Seasonality works until everyone believes it — the most obvious trade of this year is this, and here lies the real problem.' The main idea is very clear: markets generally do not reward consensus, but rather punish it.
Risky assets outside of stocks are also starting to crack. Bitcoin is currently trading around $89,460. There has been a 6.9% pullback in the last month; levels above $95,000 could not be maintained at the end of November. The market value of the leading cryptocurrency is currently around $1.78 trillion.
The Moment AI Faces Reality
A more fundamental concern arises in the artificial intelligence sector, which has brought about a $30 trillion bull market in the S&P 500 index over the last three years.
According to Bloomberg, doubts are rapidly increasing: the recent wave of sales at Nvidia, the sharp decline following Oracle's higher-than-expected artificial intelligence spending, and the deteriorating investor sentiment around OpenAI-related companies are noteworthy. Jim Morrow, CEO of Callodine Capital Management, commented: 'We are now at a stage in the investment cycle where talk has turned into action. There was a good narrative until now, but we are now actually putting money on the table, and we will see whether the investments will really yield a return.'
The cost burden is staggering. Alphabet, Microsoft, Amazon, and Meta are expected to spend over $400 billion in total on data center investments over the next 12 months. The combined depreciation expenses of these companies are expected to rise from $10 billion at the end of 2023 to $30 billion by the end of 2026.
According to a Teneo study reported by the Wall Street Journal, less than half of the current artificial intelligence projects are providing returns that exceed their costs. Nevertheless, 68% of CEOs plan to increase artificial intelligence spending in 2026. The survey results are striking: while artificial intelligence is being used more efficiently in marketing and customer service departments, security, legal, and human resources applications are lagging behind.
There are also significant differences in expectations: 53% of institutional investors expect returns to materialize within six months, while 84% of large company CEOs believe it will take longer.
Hopes Regarding Optimism
Still, comparisons to the dot-com bubble may be exaggerated. The Nasdaq 100 index is currently trading at 26 times the estimated profit. This is significantly lower than the multiple of over 80 at the peak of the bubble in 2000. The stocks of Nvidia, Alphabet, and Microsoft are also priced at less than 30 times their earnings.
So what does the past say? According to the financial newsletter The Kobeissi Letter, the last two weeks of December have been the best weeks for stocks in the last 75 years. The S&P 500 could rise to 7,000 by the end of the year.
In the short term, the seasonal effect and Fear of Missing Out (FOMO) may continue to support the markets. However, as we approach 2026, the key determining factor will be whether artificial intelligence investments truly yield results. Time will tell.
