Wall Street between AI and Santa Claus Rally: risk or opportunity

As 2025 draws to a close, Wall Street finds itself caught between two opposing forces: growing doubts about the trade in Artificial Intelligence (AI) that has fuelled this year's gains, and the historically reliable seasonal patterns that have driven markets higher in December for nearly a century.

The tension has left investors debating whether to chase the year-end rally or prepare for a possible retracement.

As the saying goes, "Crowded trades don't give easy money." The so-called Santa Claus rally, covering the last five trading days of December and the first two of January, has generated gains 79% of the time since 1929, with an average return of 1.6%.

In the past eight years, the decline has occurred only once. Nevertheless, sceptics argue that this pattern has become too notorious for its own good. The central argument is simple: the markets punish consensus, not reward it.

Risk assets beyond equities are also showing cracks. Bitcoin is trading at around $89,460, down 6.9% over the past month after failing to maintain levels above $95,000 in late November. The cryptocurrency's market capitalisation now stands at about $1.78 trillion.

The Moment of Truth for AI

The most fundamental concern lies in the AI sector, which has led the S&P 500's $30 trillion run upwards over the past three years. According to Bloomberg, signs of scepticism are growing: from the recent Nvidia sell-off to Oracle's collapse after reporting higher-than-expected AI spending, to the worsening sentiment around OpenAI-related companies.

"We're at the stage in the cycle where it's getting serious," commented Jim Morrow, CEO of Callodine Capital Management. "It's been a good story, but at this point we are betting to see if the returns on investment will actually be good."

The cost burden is staggering. Alphabet, Microsoft, Amazon and Meta expect to spend over $400 billion on data centres over the next 12 months. Their combined amortisation expenses are set to triple from around $10 billion at the end of 2023 to $30 billion by the end of 2026.

A Teneo survey cited by the Wall Street Journal revealed that less than half of current AI projects have generated returns in excess of costs. However, 68% of CEOs plan to increase AI spending in 2026.

The survey showed that marketing and customer service were the most productive uses of AI, while applications in security, legal and human resources lagged behind. There is also a gap in expectations: 53% of institutional investors expect returns within six months, while 84% of CEOs of large companies believe it will take longer.

The Reasons for Optimism

Nevertheless, comparisons with the dot-com collapse may be exaggerated. The Nasdaq 100 is currently trading at 26 times projected earnings, far below the multiple of over 80 seen at the height of the 2000 bubble.

Nvidia, Alphabet and Microsoft are all trading at less than 30 times earnings. Moreover, history favours the optimists. According to the financial newsletter The Kobeissi Letter, the last two weeks of December were the best weeks for stocks in 75 years, with the S&P 500 potentially reaching 7.000 by the end of the year.

In the short term, seasonal strength and FOMO may continue to support markets. But looking ahead to 2026, the key variable that will determine the direction of the market will be whether or not investments in AI produce real returns.