A group of 42 artificial intelligence (AI) stocks has generated 67% of the S&P 500's price returns since January 2024, exposing the benchmark index's growing dependence on a relatively small cluster of companies.

Figures shared by The Kobeissi Letter on October 11, 2026, and attributed to Bloomberg and JPMorgan, show that these companies also accounted for 58% of the index's earnings growth and 81% of combined capital expenditure (CapEx) and research and development (R&D) growth over the reported period.

The S&P 500, composed of roughly 500 large-cap U.S. companies, is the most widely followed measure of American equity performance, which is why its reliance on a narrow set of constituents has become a recurring focus for market observers. Because the index weights members by market capitalization, the largest companies exert an outsized pull on its overall level, and because many index funds and exchange-traded funds are built to track it, that pull extends directly into a broad range of investor portfolios.

"The gap between AI names and the rest of the market is historic: Since the start of 2024, a basket of 42 AI stocks has generated 67% of the S&P 500's price return. In other words, those names have contributed +22 percentage points to the index's +33% gain over this period."
— The Kobeissi Letter (@KobeissiLetter), October 11, 2026

Per the data, the remaining 458 companies in the index contributed 11 percentage points to price returns, which exclude dividends, compared with 22 percentage points from the AI cohort. The disparity means that while the broader index advanced 33% over the period, most constituents accounted for only about a third of that gain. The figures reveal how both market performance and corporate spending have become increasingly concentrated among technology-related businesses.

AI Stocks Dominate Returns and Corporate Spending

The concentration extends beyond share prices. The 42 companies accounted for more than half of earnings growth and over four-fifths of additional spending growth across the index. Much of this outlay supports data centers, advanced semiconductors and the computing infrastructure required to develop and operate AI systems, reflecting the computing power needed to train and run large AI models. Research and development commitments further widen the gap between these companies and the broader market.

A separate report from JPMorgan Private Bank's Eye on the Market Outlook 2026 found that 42 AI-related companies represented approximately 65% to 75% of S&P 500 earnings, revenues and capital spending since ChatGPT launched in November 2022. That analysis covers a longer time frame and uses different measurements from the figures shared by The Kobeissi Letter, but both datasets point to the substantial role these companies now play in corporate performance.

Recent earnings figures reinforce the pattern. JPMorgan Asset Management reported in August 2026 that just 10 companies accounted for 77% of expected second-quarter earnings growth. Semiconductor companies drove much of that growth, while heavy infrastructure spending continued to pressure profitability at some major cloud and technology businesses.

Narrow Market Breadth Leaves the S&P 500 Dependent on AI Leaders

The concentration matters because the S&P 500 represents hundreds of companies, yet its overall performance can hinge on a limited number of large constituents. Market breadth — the share of stocks participating in an index advance — is a widely watched gauge of how broad-based a rally is. Reuters reported on October 9, 2026, that the U.S. bull market remained largely driven by AI-related investments. The index had gained 117% since its October 2022 low, although rising Treasury yields, interest rates and concentration concerns threatened the rally.

The Kobeissi Letter also noted that only 29% of stocks had outperformed the benchmark over the previous three years. This limited market breadth indicates that index-level gains have not translated into comparable performance across most constituents.

The figures do not establish whether current valuations are justified, nor do they predict future returns. They do, however, identify a measurable dependence on a narrow group of companies for price performance, earnings expansion and corporate spending. The central question remains whether continued AI investment will translate into sufficient revenue and productivity gains to support sustained business performance across the sector. Quarterly earnings and capital spending updates from the largest constituents, together with breadth statistics such as the share of stocks beating the index, offer measurable checkpoints for that question.

This article is based on reporting originally published on Blockonomi.