Episode summary
The S&P 500 Is Just 50 Stocks. The Nasdaq Is a 67% AI Bet. What Else Is Hiding in Your Funds?
Excess Returns
· · 1 hr 11 min · Hosts: Matt Zeigler, Justin Carbonneau · Guests: Kai Wu
Summary
Kai Wu argues that investors often own far more of the same AI trade than fund labels suggest. Using Sparkline Analytics, he shows that the S&P 500’s apparent breadth masks heavy concentration: roughly 49% to 50% of the index is tied directly to AI infrastructure, and the Nasdaq is even more exposed at about 67%. The point is not that diversification is fake, but that market-cap weighting can quietly turn a 500-stock index into a bet on a handful of megacap tech and semiconductor names.
He distinguishes between AI infrastructure, AI early adopters, and AI laggards. The infrastructure layer has benefited most so far, but Kai argues history suggests the long-term value often migrates to the users of a new technology rather than the builders. That makes the next phase of the cycle about identifying early adopters, not just buying more AI exposure.
The conversation also shows how overlap can persist across ETFs, emerging markets, and thematic funds. Even emerging markets now carry meaningful AI exposure through holdings like TSMC, Samsung, and SK Hynix, while adding value, equal-weight, or small-cap funds may reduce infrastructure exposure but increase laggard exposure. Kai’s core message is that investors should inspect factor exposure, capex intensity, and holdings overlap before assuming they are diversified.
Key takeaways
- Kai Wu says about half of the S&P 500 is now directly tied to AI infrastructure, despite the index containing 500 names.
- He says the Nasdaq has roughly 67% exposure to the same AI infrastructure trade, making it more concentrated than many investors realize.
- Kai argues that adding dedicated AI ETFs on top of SPY or QQQ usually increases overlap rather than true diversification.
- He says emerging markets are no longer a clean diversifier because large EM weights like TSMC, Samsung, and SK Hynix are now tied to the AI buildout.
- Kai says the investment question is shifting from infrastructure to early adopters, because long-run value usually accrues to the users of a technology, not the builders.
Notable quotes
“67 or twothirds is in NASDAQ exposure to the same same kinds”
“It's always, you know, ultimately gone to the consumers”
“the true number of diversification is 50, which is not bad”
How the episode opens
You're watching Excess Returns, the channel that makes complex investing ideas simple enough to actually use where better questions lead to better decisions. I'm Matt Ziggler. Justin Carbano is my co-host today. And if you want AI beef, then bring the ruckus. Kai Woo from Spark Lane has a new tool for us to fuss with. I think the Wuang would approve of that. How you doing, Kai?
Doing good. And uh the Wuang does approve. You know, this is the way we do it.
We got a new tool to fuss with. And I say this with genuine excitement, not just because you're out of the co-host seat and we get to interview this time about this tool. Everybody I've shown this tool to a number of allocators, family offices, people like that immediately falls in love with this tool and thinks and everyone's been talking about AI and where it's shows in the portfolio, but it was really hard to find.
And here you come with a way to say, "Hey, let me show you. So what did you build? Explain what this is first for people who are encountering it the first time.
Yeah. So I mean if you step back I you know I'm a quant I build portfolios based on intangible value and have been you know doing research papers over two dozen research papers over the past few years. Many on the theme of AI others on trade globalization innovation human capital etc.
And so over the course of that and you know on the many videos I've done with you guys I've collected some really interesting data points um on on companies and their intangible capital. So the idea with this with this platform was can I provide a way to then expose this in an actionable way to the end user right like um you know so many times I can't tell you how many times I'll publish a paper on say AI and then folks will reach out being like oh that's really cool but like how does this help me right because like they don't have access to the data at their fingertips.
From the episode’s captions: roughly the first two minutes, lightly cleaned. Keep watching on YouTube →
Topics
- ai exposure overlap
- etf look-through
- factor bets
- ai infrastructure vs adopters
- portfolio diversification
Companies and tickers mentioned
- Sparkline Capital
Watch the episode
Summary, takeaways and quote selection by Octagon. Quotes are short excerpts; the full conversation is in the episode.
More from Excess Returns
-
We Asked Fidelity’s Director of Global Macro Who Wins If AI Spending Slows -
An AI Bubble Could Hit Stocks. Inflation Could Hit Bonds. How Do You Diversify and Keep Both? -
Tech Is Priced for a Boom. 86% of the Economy Grew 0.7%. What If AI Slows? -
Tech Is the Cheapest Since 2022. Walmart Is 46 Times Earnings. Where Is the Bubble? -
This $1.5 Trillion Manager Expects More Fed Hikes. So Why Is He Still Bullish? -
AI Stocks Cracked. Margins Could Be Peaking. So Why Is a Liquidity Expert Watching the Debt Wall?
Excess Returns, this episode and its artwork belong to Excess Returns. Octagon is not affiliated with the show; this page is an independent summary.
