249. Thematic Investing in 2026: AI, Defense, Infrastructure, and the Next Phase of Market Transformation
Web title: Thematic Investing in 2026, AI, Defense & Infrastructure
Episode Description:
Thematic investing is increasingly shaping how investors interpret markets heading into 2026, as artificial intelligence, geopolitical fragmentation, and infrastructure constraints intersect across the global economy.
Jay Jacobs, Head of U.S. Equity ETFs at BlackRock, joins Oscar to discuss why mega forces are becoming harder to ignore—and harder to diversify away from—than in past market cycles. Their conversation explores how AI investing is evolving from a growth narrative into one focused on usage intensity, how national security considerations are reshaping the definition of defense, and why physical infrastructure is emerging as a critical market constraint.
Key insights include:
Why thematic investing is gaining relevance alongside sector and style frameworks
How AI usage intensity reframes the AI investment conversation
Where infrastructure and energy constraints may influence adoption timelines
How geopolitical fragmentation is expanding the definition of defense
Why overlapping mega forces may shape market outcomes into 2026
Thematic investing, AI investing, Capital markets, Infrastructure, Megaforces, Stock market trends, Geopolitical fragmentation, Defense spending
Sources: iShares Thematic Outlook, 2026
Written Disclosures In Episode Description:
This content is for informational purposes only and is not an offer or a solicitation. Reliance upon information in this material is at the sole discretion of the listener. Reference to any company or investment strategy mentioned is for illustrative purposes only and not investment advice. For full disclosures, visit blackrock.com/corporate/compliance/bid-disclosures.
<<TRANSCRIPT>>
Oscar Pulido: Artificial intelligence is no longer just a growth story, it's a depth of usage story, one that's colliding with national security. Energy constraints and real-world infrastructure limits. At the same time, defense spending is evolving fast. Global infrastructure needs are exploding, and new ways of accessing markets from private capital to tokenization are starting to blur the traditional lines of investing.
So, after several years where themes didn't just influence markets, but define them, investors are heading into 2026 asking a bigger question. Are we still early in these transformations or has the big investment opportunity passed?
Welcome to The Bid where we break down what's happening in the markets and explore the forces shaping the economy and finance. I'm Oscar Pulido
Today, Jay Jacobs, head of US Equity ETFs at BlackRock joins us to walk through the big themes for 2026. We'll talk about why themes continue to punch above their weight in markets. What investors may be missing about AI, defense and infrastructure, and how the way we invest in these themes could look very different in the years ahead.
Jay, thank you so much for joining us on The Bid.
Jay Jacobs: Thanks for having me, Oscar.
Oscar Pulido: Well, Jay, you and I both live in the New York City area where the weather has been quite frigid Recently, however, the topic of thematic investing has been red hot. Some might even say white hot. And I'm curious, as you think about 2026, do you feel like themes are driving the market in a very different way as compared to maybe the last decade?
Jay Jacobs: I think over the last decade, a lot of portfolio managers really looked at the world through the lens of sector investing and style investing. growth over value tech stocks, over cyclical industrials, you name it. That was a helpful framework for thinking about the risks and opportunities in the market.
But as we've seen the emergence of mega forces like artificial intelligence and geopolitical fragmentation and demographic shifts around the world, it's really proven that there's another way to look at your portfolio to capture some of those biggest risks and opportunities, which is thematic investing.
We saw this last year when we had a big sell off in January related to the emergence of Deep Seek, That was really an AI related impact on the markets. We've also seen it with things like tariffs and geopolitical fragmentation that has shown that there can be winners and losers related to the theme of geopolitical fragmentation.
So, increasingly we're seeing people take this thematic lens as a way of looking at the market and the way of looking at their portfolios and the risks and opportunities. This is evidenced by, last year was a record setting year for the amount of inflows into thematic ETFs, showing that people are using these tools as a way to adjust those exposures.
Oscar Pulido: And what is it that makes these themes? You mentioned AI, you mentioned geopolitical fragmentation. What is it about this environment that makes them harder to ignore and then as a result, potentially hard to diversify away from?
Jay Jacobs: I think part of it's just the power of these themes. When we talk about something like artificial intelligence, the scale of this opportunity is, $10 - 15 trillion according to some estimates, you look at geopolitical fragmentation, it could completely change the way that supply chains have been built over the last 50 plus years. So, the scale of these mega forces is massive.
But the other piece of it is these mega forces are not happening in isolation. In fact, many of them are converging. Artificial intelligence is not just some interesting new technology, it could be the future of many economies around the world. And so, you see a lot of economic competition arising around artificial intelligence. frictions around things like semiconductors and materials that are inputs into artificial intelligence from supply chains around the world.
You're seeing an increasing amount of geopolitical risk around the world, which is, not maybe entirely driven by artificial intelligence, but at least maybe somewhat tangential to it. So, these mega forces are not happening in their own lanes. They are converging into one highway of change. And to your point, that means they're really impactful to portfolios, but they also make them a little bit harder to diversify across them.
Oscar Pulido: I'm recalling a discussion that you and I had, I think it was last year when we were talking about AI and that investing in AI is not just investing in technology. There can be a number of other sectors that it impacts. So, perhaps it goes to your point that these mega forces are not narrow. They're actually quite broad, and then when they start to intersect with each other, the impact on markets is even bigger.
Jay Jacobs: That's absolutely right. There's going to be entire segments of the market that are changed by artificial intelligence, which may be.
People never thought of as technology stocks, but you could think about some of the important metals that go into building a data center. You could think about power and the importance that plays in powering data centers. so suddenly you have really an economy wide impact because of artificial intelligence, not to mention the other mega forces that are having economic wide impacts as well.
Oscar Pulido: Let's stick with AI and you've pointed to how much evolution is going on in the artificial intelligence space. Let's talk about usage intensity there. There's a term here called token consumption, perhaps you could help us understand what that means and why is that a more meaningful signal for investors right now?
Jay Jacobs: Tokens are really a kind of unit of computing. And so, you could think about it like oil is for a car, oil is the power, oil is the energy that turn into gasoline that can power our car. So you can go so many miles and in many ways oil was really the thing that powered the economy in the 20th century.
Going forward, you could see tokens, these units of compute really being the commodity that powers the economy in the 21st century and beyond. So, one of the ways to measure the growth of artificial intelligence is to look at how much tokens are being consumed by artificial intelligence. And if you look at the last year, some large language model platforms have shown that they have had token usage grow between 15 to 20 x.
Often in the markets, you talk about something doubling or tripling as being a really big deal. 15 to 20 X means we're seeing a massive amount of growth in demand for AI computing power. Why is that? many of us are just using artificial intelligence more. We're using it in our everyday, maybe it's on our desktops, maybe we're incorporating it into our, into our work or our everyday lives.
But beyond that, we're seeing AI take on more complicated tasks, which require more tokens. So, if you're just chatting in an AI chat bot and saying, help me come up with a recipe for a, for a dinner next week, maybe that uses a small amount of tokens. But if you are using artificial intelligence to power an autonomous vehicle that could use a hundred thousand to a million times more tokens than that simple AI chat.
If you're using artificial intelligence to do drug discovery for a pharmaceutical company, that could be even more than a million times the amount of tokens. So, what we're, as we're seeing AI get more sophisticated, and we're seeing it being used in a lot of different ways, that token consumption is really accelerating to be able to support this incredible amount of AI need and demand in the economy.
Oscar Pulido: And Jay, I like the analogy that you made between oil and token. As oil consumption went up during a prior industrial revolution, it was a signal of people driving more and using automobiles more. And what you're saying as token consumption is going up, that signifies more people using AI or using AI for more complicated tasks.
There is, however, some growing anxiety about how much money is being poured into artificial intelligence. Again, it touches a number of different things. We're talking about energy, we're talking about data centers, we're talking about chips, and so of course the question comes up about a possible AI bubble. How do you separate healthy long-term investments from genuine excess in this space?
Jay Jacobs: So, I think there's a couple of ways to look at that question. the first is, whenever there's mention of a bubble or high valuations in a specific part of the market, you do have to ask, is it justified? Is the growth there to justify high valuations? And as we just talked about, the growth rate of AI tokens suggest that AI is growing very quickly.
But a couple of other ways to look at it, one way is to see, we spent around half a trillion dollars last year on AI capital expenditures.
But if you do compare it historically to other periods where you've seen intense investment in the United States to stimulate a revolution, AI doesn't quite top the lists yet. For example, if we go back to the 1910s where you saw a tremendous amount of spending around the automobile revolution to manufacture automobiles, build roads, that was a significantly higher percentage of GDP, at the time, than what we see with AI today.
And I don't think anyone's looking back and saying, we overinvested in vehicle transportation. It's used every day by millions of people, of course. So, when you look at it through that lens, it suggests that yes, AI's a big deal or it's spending a lot of money to build out this capital expenditure. But in the historical context, it's not actually the biggest investment we've seen in the United States.
The other way to look at it is really trying to understand, are these companies growing earnings? Is this spending justified because they're being more profitable, or is it just money spent and going out the door? What we found with AI companies last year, was they grew earnings tremendously, actually, to the point where they grew earnings faster than the price of these stocks appreciated.
Valuations for AI related companies in the S&P 500 actually came down last year because the earnings growth was happening faster than price appreciation. You can't say that about the rest of the S&P 500. The other companies that were not related to AI saw decent performance last year. About half of that was due to earnings, and half of that was due to valuations getting stretched.
So, from our lens, it suggests that AI companies are doing a pretty good job of being profitable. Growing their profitability through their investments and that in the context of history, these investments are still very well justified for the size of the opportunity of artificial intelligence.
Oscar Pulido: And since you mentioned history, there's often a parallel that people try to draw to the late nineties when we saw the technology sector doing very well, but I think the difference back then was that prices were going up a lot and earnings weren't. I think what you're saying here is that within the AI space, which again is not just technology, but if we try and draw that parallel earnings have been growing quite fast and prices have either been keeping up or in some cases not keeping up. So, you actually have better valuations?
Jay Jacobs: That's exactly right. The 1990s was a lot more speculative in the growth of these companies than the valuations of these companies in ai In the 21st century, we're really seeing a lot more of it justified with real profitability of these companies.
Oscar Pulido: We talked about how mega forces can intersect, so let's go to this mega force of geopolitical fragmentation.
And that means that national security and topics like defense are suddenly back in the investment conversation. In a big way. The Bid was recently at Davos, at the World Economic Forum, and that certainly featured very prominently in some of our conversations with, Tom Donilon, who was one of our guests.
What does this entail and what's driving that shift and why do you think markets were slow to price in this theme, but maybe are starting to pay a little bit more attention?
Jay Jacobs: for many years following World War II, you really saw more global integration. You saw the growth of global shipping. You saw global passenger travel through airlines. You saw the growth of global data through the internet, and that really tied countries closer and closer together. What we've seen recently is more challenges to this. You've seen supply chains being rewired. A lot of countries focused on how can they build more within their own borders, whether that means more manufacturing, hiring more skilled workers in critical industries, being protective of critical industries like artificial intelligence.
And so that has somewhat embodied the economy that we are now entering, which is a little bit less globally integrated and a little bit more, self-sufficient with countries trying to provide all of the services and manufacturing that they need within that own country. And as a result or maybe adjacent to that, we're seeing more conflict, more potential for conflict around the world.
Oscar Pulido: And is the investment opportunity in defense changing at all? it feels like we hear a lot about cyber and space and autonomous, type of defense systems and AI sometimes comes up in this conversation as well. In the past, I think we talked about tanks and ships. So how is that investment opportunity set evolving as we think about going forward?
Jay Jacobs: It is shifting. For a long time there's been a sub-sector called aerospace and defense, which consists of a lot of major US companies that lead in the building of aircraft and other defense systems. But what we're seeing is that the definition of defense is starting to change pretty, dramatically.
One is it's not just a US story. Some of the biggest regions for defense spending growth are Europe, the Middle East, where you have local defense companies that are seeing a lot of new orders, a lot of growth.
The second piece is a lot of things related to defense don't really fit into that aerospace and defense category anymore. Software and AI companies related to defense, critical metals and minerals, infrastructure that's critical to defense, new growth areas like space. All of this is starting to encompass a broader definition of what defense means that extends beyond more 20th century definitions.
Oscar Pulido: You mentioned infrastructure. Again, this is one of these themes that you and I, have talked about when we talked about the physical economy. And I think this is a term that maybe has been regarded in the past as a bit slow and boring, and we think about bridges and tunnels, now it seems that infrastructure is really central to economic growth and AI and that it's at the center of geopolitical discussions as well. , What's changed?
Jay Jacobs: A few things have changed with Infrastructure. One is simply a lot of the infrastructure that's been built in the United States is aging.
A lot of it was from the sixties. We're now in 2026. It's past its ex, its, expected lifespan and it needs to be rebuilt. The second is the demands on infrastructure have really shifted. Yes, we've had population growth. Yes, we've had economic growth, but different parts of the economy are putting new strains on infrastructure.
For example, you look at artificial intelligence, it requires a tremendous amount of power. And so having a power grid that can supply this new power demand growth, not just with the right fuels, but with the power generation, with the power distribution across the grid, all of that could require a lot of capital expenditures to really build infrastructure that makes sense, for this AI revolution.
And then finally you're really seeing a shift in how that is being funded. Historically, governments have really taken on the lion's share of that funding for infrastructure, and what we're seeing is as these infrastructure demands continue to grow because of things like artificial intelligence and as government debts and continue to expand, a lot of that infrastructure spending and funding is shifting to the private sector.
Oscar Pulido: Jay, at times when we talk about infrastructure, and I think you alluded to this, it's also about trying to meet the energy and power demands of a more modern economy. And there are in fact constraints to, energy and power, especially with things like artificial intelligence continuing to develop, we're seeing that, especially in the U.S. So how much do physical realities matter in shaping the investment outcomes from here?
Jay Jacobs: I mean, the physical realities make a tremendous impact on the growth of artificial intelligence. we like to think of AI as just happening in the cloud somewhere out there in the ether, but there's actual real physical realities, which is you have to have a data center that has to be built, that has to be permitted, that has to be supplied with power, and some of those things can take years.
Not just to build a new power plant, not just to build the new wires that connect that power plant to the data center, but frankly, to work its way through communities and regulation to be able to bring out those power plants. So as we continue to scale ai, as we continue to see token usage grow, you really need to build out some of that physical infrastructure behind it to be able to support those ambitions.
Oscar Pulido: I think that's a great point. When we talk about artificial intelligence, it seems very futuristic and something that's at our fingertips that is making our life simpler. But behind the scenes, there's a lot of physical infrastructure and demand on the physical economy that it's creating.
Jay, AI feels like something we talk about a lot, but it also feels like something that is very much in its infancy in terms of the story. So, what's next for the AI story as it's such a big theme and what's next for investors more broadly that they need to consider?
Jay Jacobs: One of the keys that investors need to look at for 2026 is this idea of the broadening out of the AI story. The Mag seven - the Magnificent seven stocks - the large mega cap tech stocks in the United States have garnered so much attention when it comes to AI. It's not all about them. There's other parts of this value chain that are really going to matter. We talked about the power. How are these data centers going to get the power they need to be able to operate within the chip space?
It's not just about these high-powered GPUs that are training large language models. It's about storage. How do you store more data when data becomes more valuable in this AI landscape? You could even look beyond the AI value chain and just look at companies that are adopting ai. What are some of the companies that might be able to accelerate their earnings growth?
Because they're using AI very intelligently. So, as we, as we look at this AI opportunity, I think people have to look at the broadening of the value chain. They have to look at companies that are adopting ai. And I think another aspect that we can't ignore for 2026 is. There might be some really big IPOs this year in terms of AI related companies coming out, which would further broaden the opportunity set. So, 2026, headline broadening of the AI story,
Oscar Pulido: Right, more companies, more sectors that we will be talking about that are interrelated with this theme. In fact, you started by saying, we used to talk about markets more through the lens of large caps and sectors and regions, and we still do that to a large extent, but certainly themes have become part of conversation in a much bigger way. And Jay, you always do a great job of helping us understand, where we are in the cycle as it comes to themes. Thank you for sharing some of your insights, and thanks for doing it here on the bid.
Jay Jacobs: It's always a pleasure to be on here with you, Oscar.
Oscar Pulido: Thanks for listening to this episode of The Bid. If you enjoyed this episode, check out episode 245, a Stock Pickers guide to 2026 where Carrie King tells us about her equity picks for this year.
<<SPOKEN DISCLOSURES>>
This content is for informational purposes only and is not an offer or a solicitation. Reliance upon information in this material is at the sole discretion of the listener. Reference to the names of each company mentioned is merely for explaining the investment strategy and should not be construed as investment advice or recommendation. For full disclosures, visit blackrock.com/corporate/compliance/bid-disclosures.
MKTG0226-5163126-EXP0227
