Most investors holding S&P 500index funds believe they already have solid exposure to the most profitable names in the artificial intelligence trade. The tech sector has returned roughly 115% over five years through the Technology Select Sector SPDR ETF (XLK), dwarfing the index’s 65% gain. 

But even that gap understates what a handful of AI companies have been doing beneath the surface of the broader market in recent quarters. The largest AI profits are not distributed evenly across the index, and your portfolio may be capturing far less upside than you realize. 

Three companies have been compounding at rates that could make the S&P 500 look pedestrian over the next half-decade, and each one occupies a different layer. You may already own all three inside your index fund, but at weightings so small that you barely benefit from their momentum.

Tech stocks have been outrunning the S&P 500, and 3 names lead the charge

The State Street Technology Select Sector SPDR ETF (XLK) tracks tech stocks inside the S&P 500 and has doubled in value over the past five years. The S&P 500 itself returned 65% over that same stretch, meaning tech investors earned roughly 35 percentage points more than the broader index did. 

Three stocks account for a significant portion of that outperformance, and each one ties directly to the AI infrastructure buildout, Motley Fool analysis shows. Those three names are Broadcom (AVGO), Micron Technology (MU), and Alphabet (GOOGL), and each plays a distinct role in the AI economy. 

Broadcom builds custom AI chips for hyperscalers, Micron supplies the memory powering every AI processor, and Alphabet operates Google Cloud, one of the fastest-growing enterprise platforms.

Broadcom’s custom chip business carves out territory Nvidia does not occupy

When you think of AI semiconductors, Nvidia probably comes to mind first as the dominant supplier of general-purpose graphics processing units. But Broadcom has built a separate empire in custom-designed chips called application-specific integrated circuits (ASICs) that serve hyperscaler clients. 

Alphabet and Meta Platforms use Nvidia for general workloads but turn to Broadcom when they need chips engineered to their own specifications. Broadcom’s AI semiconductor revenue reached $6.5 billion in Q4 of fiscal 2025, representing a 74% increase year over year, the company’s SEC filing shows. 

“We see the momentum continuing in Q1 and expect AI semiconductor revenue to double year-over-year to $8.2 billion,” CEO Hock Tan told investors on the earnings call, according to Barchart. The company’s total AI backlog now exceeds $73 billion, with delivery expected over the next 18 months, providing unusual revenue visibility.

In Q1 of fiscal 2026, Broadcom followed through on that guidance, with AI chip revenue surging 106% year over year to $8.4 billion.  The company also secured long-term supply agreements with Google and Anthropic for custom chip development extending through 2031.

J.P. Morgan has named Broadcom its top semiconductor pick, projecting $55 billion to $60 billion in AI-related revenue for fiscal 2026.

Broadcom CEO Hock Tan highlights surging AI chip revenue as custom ASIC demand from Alphabet and Meta Platforms accelerates growth beyond Nvidia dominance.

Dragos Condrea/Getty Images

Micron’s memory chips are hidden engine inside AI processors sold today

Every AI chip sold by Nvidia, Broadcom, or AMD contains high-bandwidth memory components manufactured by companies including Micron, Samsung, and SK Hynix. When any of those chipmakers sells an AI accelerator, it automatically creates direct demand for Micron’s products embedded inside the silicon. 

Think of Micron as the essential building block inside the building block, earning revenue from nearly every AI chip transaction on the market today. Micron reported fiscal Q2 2026 revenue of $23.86 billion, nearly triple the $8.05 billion from the same quarter one year earlier, according to its SEC filing. 

Non-GAAP earnings per share reached $12.20, far exceeding Wall Street’s consensus estimate of $9.31, CNBC reported in March.  The company guided fiscal Q3 revenue to $33.5 billion at the midpoint, representing roughly a 40% sequential jump from the prior quarter.

“The step-up in our results and outlook are the outcome of an increase in memory demand driven by AI, structural supply constraints, and Micron’s strong execution across the board,” said President and CEO Sanjay Mehrotra.

“Micron set new records across revenue, gross margin, EPS, and free cash flow in fiscal Q2,” Mehrotra said during the earnings call. Micron also shut down its consumer Crucial brand to reallocate all manufacturing capacity toward higher-margin AI and data center products. 

Micron Chief Business Officer Sumit Sadana said bluntly that “our supply is nowhere close to being able to meet the demand we see.” Micron’s forward price-to-earnings ratio sits around 7, compared to Nvidia’s 23, making it one of the most attractively valued AI stocks available today. 

Supply remains constrained, and management has indicated that the memory shortage will likely persist well beyond 2026, keeping pricing power in the company’s hands. Investors who dismissed Micron as a cyclical commodity play, as Global Banking & Finance Review indicated, are now watching it post growth numbers typically reserved for high-multiple software companies.

Alphabet’s Google Cloud division becomes the financial engine 

Google Cloud has transformed from a money-losing division into Alphabet’s fastest-growing and most strategically important business segment over recent years. In Q4 of 2025, Google Cloud revenue surged 48% year over year to $17.7 billion, driven by demand for AI infrastructure and enterprise solutions. 

The division’s backlog more than doubled year over year and grew 55% sequentially, reaching $240 billion at the end of the fourth quarter, Alphabet’s SEC filing shows. Cloud is where Alphabet monetizes AI directly, and the customer roster includes OpenAI, Anthropic, and major enterprise clients across financial services and health care. 

More AI Stocks:

• Morgan Stanley sets jaw-dropping Micron price target after event
• Bank of America updates Palantir stock forecast after private meeting
• Morgan Stanley drops eye-popping Broadcom price target

Nearly 75% of Google Cloud customers have used Alphabet’s AI products, and those AI customers spend 1.8 times more on the platform. CEO Sundar Pichai told investors the Gemini AI app reached 750 million monthly active users during the most recent quarter, TechCrunch reported.

Alphabet revealed planned 2026 capital expenditures of $175 billion to $185 billion, CNBC reported, more than doubling its 2025 spend to meet surging enterprise demand. Shares are up 184% over the past five years, comfortably outpacing the S&P 500’s 65% return over that same period for long-term investors.

Do AVGO, MU, and GOOGL deserve a bigger place in your portfolio?

Owning the S&P 500 gives you exposure to all three companies, but the math may not work in your favor at standard index weightings today. 

If these three stocks continue growing at their current pace, your index fund captures only a fraction of the gains they are generating individually. Investors who are comfortable with individual stock selection may want to evaluate whether increasing direct exposure makes strategic sense for their situation.

Key considerations before adjusting your allocation

• Review your current portfolio allocation to understand how much direct AI exposure you actually hold through your index fund positions today.
• Consider whether individual position sizing in Broadcom, Micron, or Alphabet aligns with your personal risk tolerance and your investment time horizon.
• Evaluate each company’s valuation metrics independently, rather than relying on broad index rebalancing to do the work of portfolio construction for you.
• Keep in mind that concentrated positions carry a higher risk, and past performance does not guarantee that future returns will follow the same trajectory.

The risks every AI investor should weigh 

Broadcom faces gross margin pressure as custom chip sales are less profitable than its traditional products, according to GuruFocus, and management has already warned investors about margin compression.

Micron has historically been a boom-and-bust cyclical stock, CNBC noted. Current memory prices could reverse if supply catches up to demand faster than projected. 

Alphabet’s planned $175 billion to $185 billion capital expenditure program could weigh on free cash flow if AI monetization disappoints investor expectations, CNBC reported. Tariff uncertainty also looms over the entire semiconductor supply chain, and none of these companies has fully priced in potential trade disruptions. 

AI spending by hyperscalers could also decelerate if enterprise adoption disappoints or a broader economic downturn squeezes corporate technology budgets across sectors.

Diversification remains the most reliable approach to managing risk, even when individual stock stories look compelling and growth numbers seem extraordinary.

Related: S&P 500 index dividend yield hits nearly 50-year low