S&P 500 now 45% AI-driven as megacap tech dominance raises market risks

AI stocks are dominating the S&P 500, with the AI boom now pushing their weight to nearly 45% of the index’s total market capitalization. The dominance of this “AI backbone” is primarily driven by a handful of megacap leaders and a surge in infrastructure spending.

AI stocks tied to data center, semiconductor, and energy firms now account for over 40% of the S&P 500’s total value. The high concentration in a few names increases risk if AI revenue monetization fails to meet expectations.

Goldman Sachs estimates that AI infrastructure investments will account for approximately 40% of all S&P 500 earnings growth in 2026. Data center construction and AI capital expenditure have reached a structural scale, on track to hit 2% of U.S. GDP by late 2026. Analysts from Capital Economics suggest that the S&P 500 would be trading roughly 25% lower without the AI boost.

NVIDIA becomes the single most influential AI stock in early 2026

According to S&P 500 data, Nvidia is the most influential stock, with a 7% weight in the index as of March 30, 2026. NVIDIA’s stock has surpassed Apple (6.3%), Microsoft (4.6%), and Amazon (3.7%) in index influence. The top five AI companies now hold roughly 30% of the S&P 500, the highest concentration in half a century, effectively transforming the broad benchmark into a mega-cap tech fund.

The top 20 AI-related stocks account for nearly half of the index’s weight, a level that surpasses the peak of the 200 Dot-com bubble. Investors have rotated so heavily into AI infrastructure and semiconductors that other industries, like cybersecurity and enterprise software, were sidelined for much of early 2026. The narrative has shifted from growth potential to tangible monetization, meaning that a correction in just 3-4 AI mega-caps could trigger a systemic deleveraging that the other 480 stocks in the S&P 500 index would be unable to offset.

Hyperscalers’ capital expenditure cements their roles as growth drivers

Beyond chips, massive capital expenditure (CapEx) from hyperscalers like Microsoft and Alphabet (projected to spend nearly $700B collectively on AI in 2026) has cemented their roles as the market’s primary drivers of growth. AI-related firms have seen total gains of 200% since ChatGPT’s launch in 2022, while the remaining ~459 companies in the S&P averaged just 27%. That means any slowdown in AI CapEx could trigger a broad market repricing.

The “Big Four” (Amazon, Alphabet, Meta, and Microsoft) are expected to spend approximately $645-700 billion on AI infrastructure in 2026 alone, a 50-60% increase from 2025. However, achieving true portfolio diversification has become increasingly difficult as themes across industrials, energy, and technology are now all correlated to the data center buildout.

The high concentration of AI stocks has made the S&P 500 index “brittle,” even as the market shifts from blind trust to demanding proof. Investors are now scrutinizing whether this massive AI spending is translating into measurable revenue growth and margin expansion because many are “long” on AI.

There are rising fears that the obsession with AI is sidelining other industries as capital and focus are sucked away from sectors like traditional retail or healthcare. Even minor negative news can trigger outsized market drops. Analysts from Morgan Stanley and Goldman Sachs recommend shifting the focus from broad tech exposure to specific AI adopters with pricing power and infrastructure plays that bridge into the real economy, such as manufacturing and energy.

In 2025 and early 2026, top performers driving this AI trend included GE Vernova, Seagate Technology, Palantir Technologies, and Super Micro Computer. The focus has recently shifted toward companies building physical AI infrastructure, such as Lumentum, Vertiv Holdings, and Coherent, which were added to the S&P 500 on March 3, 2026. The infrastructure boom is also heavily reliant on energy, with companies like GE Vernova and NRG Energy benefiting from the demand for power in data centers.