How AI Is Taking Over Your Portfolio

Ivanna Hampton: Welcome to Investing Insights. I’m your host, Ivanna Hampton. The dominance of artificial intelligence is swelling globally. Its massive growth is weighing heavily on stock markets, pressuring bond markets, and even growing to outside proportions in emerging markets. These developments are increasing concentration risk in everyday investors’ portfolios. What should you look for, and how do you guard against this AI surge? Tom Lauricella is Morningstar’s chief global markets editor and the editor of the Smart Investor newsletter. Tom, you’re making a rare Chicago studio appearance. Welcome.

Tom Lauricella: Glad to be back here.

Hampton: Well, it’s been more than three years since the start of the AI explosion in the stock market. Are you seeing any signs of it slowing down?

Lauricella: We’re really not. The dynamics of it have changed, but there continues to be just enormous amounts of money supporting the AI infrastructure build in the stock market as well as the bond market, which we’ll talk a little bit about. But no, it continues to be a really important driver, especially in certain sectors of the market, certain industries, but it’s remarkable. I mean, it’s been over three years, and really, based on what analysts are saying, what companies are saying, it doesn’t look like it’s going to end anytime soon.

Hampton: Now you’ve written about stock market concentration and included this chart comparing semiconductors weighting against other industries like software. What does semiconductor stock weighting mean for investors’ portfolios?

Lauricella: Yeah, it’s remarkable. The idea of market concentration has been building over the last few years in the stock market. First, it was the Mag Seven, and then we just saw more of some of the AI names building and building. And so a lot of people are aware that they have a very heavy tech concentration, but what I don’t think people understand is, even within technology, how concentrated they are in semiconductors. Semiconductors are historically very cyclical. They’re even more volatile than just general tech stock.

And yet we’re at a position now where semiconductors are about 16% of the US stock market, and that’s a very high number for such a volatile industry. So investors have a lot more exposure to this traditionally cyclical, and right now some of these companies, very high valuations. There’s a lot going on here that investors may not realize in their portfolios.

Hampton: And we’re going to get into that because this is not just a US development. Emerging-market stock funds are also tilting toward semiconductor companies like Taiwan Semiconductor Manufacturing TSM. Talk about how concentration risk could overshadow the risks that are typically associated with these type of funds.

Lauricella: Sure. Yeah. When people think about emerging markets, historically it tended to be these big macroeconomic factors, currency issues, inflation, in particular, and then things like political risk. That’s tended to be what most people think about when they think about emerging markets. And over time, that, interest rates, has tended to be the big driver.

But instead, now what we’ve seen is even more concentration in emerging-market funds. We’re talking about 22% of your big emerging-market indexes. That’s a lot, a quarter of a portfolio in just a handful of companies, three or four companies. That’s a really big change for the emerging markets. A year and a half ago or so, it would’ve been about half of that.

Hampton: And what have Morningstar and outside analysts and strategists, what have they said about managing risk in emerging markets?

Lauricella: Yeah, it’s a difficult question. And Gregg Wolper had a great article looking at how emerging-market funds are handling this in different fashions. It gets back to this basic concept of how active managers, in particular, will run their funds. Even though they’re active, they pay very close attention to the benchmarks because their portfolio performance is measured not just against other funds, but against an index benchmark. So if you’re a portfolio manager and 22% of your index is in these three companies, four companies, at most, then you have a very big decision to make. If these companies continue to perform as strongly as they have, but you’re underweight, you’re going to lag, and it’s not going to look good even if you believe that these companies have run too far, too fast, the risks are too high.

So fund managers are taking a different approach. You have got some funds—Gregg identified one at T. Rowe that’s some 33% in those three names. At Fidelity, he noticed that there were two funds, emerging-market funds, one was very overweight, these names, another was underweight. And then you have an emerging-market fund at Dodge & Cox, he mentioned, that is very much underweight these names, just single digits in these companies. So there’s just different ways, and really it’s a difficult position for these fund managers. This is such a highly concentrated market environment.

Hampton: But even investors may want to peek and see what the percentages are in their emerging-market funds.

Lauricella: Yeah, I mean this is something just to look at across portfolios right now. So I mean if you just think about it, you’ve got 15% in semiconductors in the US, then you’ve got your emerging-market portfolio that you use perhaps to diversify some of that US risk, but that’s 22% in semiconductor stocks basically. So even when you look to diversify right now, you’re not getting much diversification. It’s a very challenging environment.

Hampton: Indeed. During the July 24 episode of Investing Insights, I discussed with Morningstar ETF specialist Dan Sotiroff about how AI is dominating many broad-based market ETFs, but they’re also affecting value ETFs. Can you explain?

Lauricella: Sure. So there’s another knock-on effect here, and this gets a little bit into the weeds of how indexes weight the stocks, and it gets a little complicated. We don’t need to go into details, but for example, the net result is that at the end of 2024, your typical value, some of the big value indexes, would’ve had technology stocks at about 11% of the portfolio. But now we’re seeing some key value indexes at something like 20% technology, and that doesn’t include Amazon AMZN, which is not considered a tech stock. Amazon is now the largest holding in the Russell Large Value Index at 6%.

There are a lot of value ETFs that now have something like more than a quarter of their portfolio in tech and tech-adjacent stocks. That’s not what usually people think of when they think of value. They think of financials. They might think of energy. This is a really big change. And this is actually a knock-on effect of what we are seeing in terms of the AI stocks, because as those stocks have rallied, become more highly valued, they’re pushing down Big Tech names like Apple AAPL, Microsoft MSFT, Amazon into the value indexes. It’s really interesting and again, something that I think probably a lot of investors would never think of when they think about their portfolio.

Hampton: And you’ve been covering markets for a long time, so this is something that’s catching your attention.

Lauricella: Yeah, this continues to be a very unusual environment that continues to get even more unusual. This AI infrastructure buildout, the ripples from it just keep spreading, and we’re not done yet.

Hampton: As we’re talking about the value ETFs and how it’s changing, what’s behind this trend? Why are the lines blurring between growth and value categories?

Lauricella: Yeah, again, this gets back to that question of how the indexes, actually index providers, decide what’s a value and growth stock. It’s all relative, right? There’s no just sort of bright line, and it moves around depending on valuations. It depends on relative growth. So when you have something like these hardware stocks or semiconductor stocks that have had this explosive growth in revenues because of the AI infrastructure buildout, a company like Microsoft is still growing pretty handsomely. Apple, same thing, but compared to something like the revenues that you’re seeing in Sandisk SNDK or Western Digital WDC or Nvidia NVDA, Apple and Microsoft look like value stocks just because of where they are in the rankings. Again, this is one of these weird ripples that just continues to spread.

Hampton: So the phrase, it’s all relative.

Lauricella: Yes, this is all relative. Value and growth is very relative, and it’s forcing people to rethink what do they think of when it’s a value stock? And really it’s just a list that gets divided in a certain place.

Hampton: Let’s rewind back to the May 8 episode of Investing Insights. Morningstar Investment Management’s Chief Multi-Asset Strategist Dominic Pappalardo weighed in on Big Tech’s bond-buying bench. He said then that AI debt made up a very high share—you said those same three words earlier—of the corporate bond universe. What’s the outlook for bond buying for the rest of the year and 2027? What does that spending look like?

Lauricella: Yes. And so this is yet another knock-on effect, another ripple of this buildout. These companies are spending massive, massive amounts of money to build AI infrastructure, data centers, power generation, everything that goes into the technology, and they need to fund it. And even companies as giant as Amazon and Alphabet and Meta META can no longer afford to finance this infrastructure buildout through the cash that they generate. So they’re having to issue debt to finance these projects. So the amount of bonds that are being sold that are from these hyperscalers or data center financing has exploded this year, and it’s surpassed even very high expectations earlier in the year.

And that’s changing the dynamics in the corporate bond market now, too. Most analysts that we’ve looked at expect those numbers to continue going up into next year, potentially level off as we get into ’28, but we have some time to go before this debt slows down, and we’re seeing this as far afield as the Swiss bond market and the Canadian bond market. They’re not just issuing here; they’re issuing globally because they’re issuing so much debt, they need to spread it out. And the knock-on effect there is the same as we were talking about before, which is that now, you’re starting to have less diversification benefits from the corporate bond market because now corporate bonds have a much higher weighting in technology companies than they have pretty much ever.

Hampton: You just said 2028 at this table. I’m not sure if I’m ready for that year yet.

Lauricella: Yes, it’s coming.

Hampton: Are bond yields expected to rise higher? What are market analysts and strategists telling you about that?

Lauricella: Right. So we’re in a very interesting period in the bond market at the same time here. While most people just focus on what’s happening in the stock market, long-term bond yields have been rising throughout the year. Part of this has to do with inflation worries that are spilling over from the energy price spike that was caused by the Iran war.

There’s growing concerns about fiscal deficits around the world. The US just topped 40 trillion, I believe, and it’s the same in Europe, the UK. The yields are going up in Japan. This is happening everywhere among major markets. And the competition for government debt from these hyperscalers that are very high-quality issuers is—basically there’s some what analyst will say is sort of a “crowding out mechanism.”

There’s only so much money to go around, and investors are just demanding higher yields because they have more options. And so most analysts seem to think that we’re probably close to where yields will go. However, we’ll have to see. If inflation stays high in the US and Europe, we could see long-term interest rates continue to edge a little bit higher.

Hampton: And with all this happening, Anthropic and OpenAI are expected to go public. What should investors know about these IPOs?

Lauricella: Yes. So the hits just keep coming. Anthropic, we don’t know exactly when it’s going to IPO. The talk is that’s going to be late September, October. By the time this airs, there might be an answer on that. This will likely be the largest IPO ever, surpassing SpaceX SPCX, which is currently the largest IPO by a wide margin. And these are very significant IPOs because these are the two biggest companies out there in the AI space. These are brand-new business models. There’s going to be a lot of questions about how these companies make money and their financial commitments and debt. And at the same time, these are, to some degree, going to be must-own stocks for a lot of institutional investors.

I think what investors are going to need to take a look at is, again, especially once these companies go public, what happens to their exposure? It’s likely to tilt the market even further, we don’t know by how much yet, to a technology concentration. Investors just really need to be aware of how much of their money is in technology and technology-related companies. Right now, that’s the growth engine. Seems to be supporting the economy. A lot of people say it’s all good, it’s all good. At the same time, the valuations always matter, and investors just need to be aware and be conscious of this.

Hampton: And we’re going to want to timestamp this for everyone: We are recording this episode on Tuesday, Sept. 1, just before noon, everyone. I know things can happen very quickly in the markets. Tom, I know we’ve been sprinkling what investors should do, what they should guard for throughout our conversation, but let’s button it up. What are the takeaways that investors need to keep in mind as AI extends its reach and its dominance grows?

Lauricella: It’s a very difficult and really unprecedented time. And there have been other points in time in the markets where things have been very concentrated. They still talk about an AI bubble less, I think, right now. But I think at this point, investors just mostly need to be aware of the risks, aware of the potential for volatility, and as always, put their investments into the context of their financial plan. So how much volatility can you stomach? How much drawdown? What if there is a hiccup in the AI buildout? What if there’s some other issue that happens? Really, it’s really important. It’s always important to know what your risks are, but now more than ever.

Hampton: All right. Well, everyone listening and watching, check out the link in the show notes so you can sign up for Tom’s Smart Investor newsletter. You want that, right?

Lauricella: Yes.

Hampton: Great. Tom, thank you for stopping by the table while you were visiting us in Chicago.

Lauricella: Always a pleasure.

Hampton: That wraps up this week’s episode. Thanks for making this show part of your day. A couple of reminders: Give Investing Insights five stars on Apple Podcasts to help others find the work we’re producing for you, and subscribe to Morningstar’s YouTube channel to watch new videos from our team. Thanks to senior video producer Jake VanKersen and associate multimedia editor Jess Bebel. I’m Ivanna Hampton, editorial multimedia manager at Morningstar. Take care.