Subscribe to the Alger Podcast in
Spotify |
Apple Podcasts
In this episode, Brandon Geisler, senior vice president and portfolio manager of Alger’s Mid Cap Growth Strategy, discusses how the AI opportunity is evolving from buildout to broader adoption, what that could mean for mid cap companies, and how he identifies potential opportunities.
Rudy Majid: Hello, I'm Rudy Majid and you're listening to the Alger Podcast, Investing In Growth and Change. From the infrastructure supporting AI to businesses adopting it, what could this evolving technology mean for mid cap investors? Here to discuss where we are today, what the road map ahead looks like, and how he identifies potential opportunities, is Brandon Geisler, senior vice president, and portfolio manager of Alger's Mid Cap Growth Strategy. Brandon, thank you so much for joining us.
Brandon Geisler: Pleasure to be here with you today, Rudy.
Rudy: Before we get into our initial discussion on AI adoption, I wanted to talk to you about the AI buildout. Where are we today with that?
Brandon: Yeah, I think from looking at this from various different measures, there's no doubt we're still in the infancy. We're only really four years into this buildout and this is something I think we're all going to be dealing with and working with through our lifetimes.
You know, we tend to think about 2022 as the ChatGPT moment, but if you think about what's evolved over the last four years and you look at many of the business across the world, they are just now starting to really implement AI, outside of the technology sector, and I think from our perspective, it's an exciting time to see different businesses deploy this technology and, in our view, there is going to be a lot of disruption in markets as this happens over the next several years, and having a front row seat to that in the mid cap portfolio is very exciting for us.
Rudy: Well, speaking about the mid cap portfolio, where do mid cap companies fit into all of this?
Brandon: Yeah, a lot of the attention, and rightfully so, has been centered on large cap over the last few years, but as you move away from that, you can step back and think about who's actually building these data centers. Who are the real estate agencies working with them? Who are the cooling companies? Who's providing some of the other technologies, not just the Nvidia chips or the GPUs, but the associated technologies when it comes to things like optical sensors. So, there's a tremendous amount of ecosystem around the deployment of GPUs inside data centers, and we're lucky in the mid cap space because in many cases, they're directly involved.
Rudy: Brandon, I want to pivot to the adoption side of things. There was a recent paper involving MIT FutureTech looking at the AI adoption across S&P 500 companies. I know you came across this too. So, what stood out to you in this paper?
Brandon: This particular paper attempted to categorize S&P 500 companies in their level of deploying AI, and I think what was interesting about it is when you dive into the details, obviously technology is an early adopter and greater than 70% of the technology companies have actually really taken AI and deployed it through their organization and the products they're creating. But once you step away from technology and you look at sectors like healthcare, or retail, or materials and industrials, the penetration rate in some cases is as low as zero, as far as full penetration. And I think what's exciting about this, and intuitively also it makes sense…companies that are closest to the core technology that have arguably the least amount of tech debt are in the best position, in our view, to kind of deploy AI.
And I think the exciting thing for me from my perspective is, many of the industries are on that pathway and every quarter, every year that goes by, we're going to see more and more of those industries starting to deploy the technology.
The other interesting thing I think that came out of this from an investor perspective is that in some respects, deployment of AI almost follows like the J-curve in private equity, where you've spent several quarters or even years understanding the technology, maybe having incremental employees helping you or consultancy helping you, as you're trying to figure out where it can actually improve revenues or improve operations, and it takes a period of time before you have the “aha” moment and are actually ready to deploy and include it as a core component of your service or your product, and that's when I believe you actually start seeing the benefit.
Rudy: I wanted to break up what you said into two parts, so let's start with the first part and then later unpack the J-curve. You mentioned to us that there is an adoption rate, but there seems to be a distinction between AI adoption and deep integration. So, what does AI integration really look like and can you describe that process?
Brandon: I think one of the ways to think about it is to even think about in a personal use case for all of us. Our earliest experiences were simple chat with an AI with maybe Claude or ChatGPT interaction, and you look at many of us today in different products like Meta Muse, we're now thinking increasingly at a personal level about things like agentic AI and having our own agents. I think businesses are on a similar kind of pathway. Initially, they encourage adoption internally and trial, and they're asking their employees to use the product, potentially to come up with use cases. Then they go through a progression of what are the security implications, what are the operational implications, what are the highest paybacks? One of the things that obviously we've seen in markets over the last six to nine months is, a lot of token usage by different organizations and then throttling into ensuring that, for example, the right composition of closed-end model use and open-end model use as they try and manage operating budgets. So, I think businesses are on that similar continuum that many of us are as personal users, and I think once they've kind of found the right mix of hey, we can use artificial intelligence maybe in this particular business process. These are the economic benefits of that implementation. These are the operational and security benefits and considerations, and then it becomes, I think, deeply embedded in process, and that's when you finally see the inflection in the business, and I think, in certain industries, they may have identified 30 or 40 projects, for example, that are applicable, but they're at the current moment maybe only addressing 10% of those because they have the highest immediate payback and the easiest to implement, but the tail is long and growing.
Rudy: You mentioned that tech companies are further ahead in their adoption than non-tech companies. What does that tell you about where the next opportunities could emerge in the mid cap space?
Brandon: So, as we think about the mid cap space, for me what's exciting is that really is it's part of this broadening out story for artificial intelligence, where much of the storyline initially has been around data centers, has been around semiconductors, has been around power, and that will still be an important storyline undoubtedly, but what we're going to start to see, I believe over the next one, two, three, four, five years is a lot of disruption in other business models as well across the economy. And from my perspective, ferreting through good ideas in the mid cap space that take advantage of this implementation is an exciting road ahead.
Rudy: Well, I'm definitely going to want to talk to you about your process of picking those businesses, but I do want to circle back on the second major point you brought up, which was the J-curve. So, you were mentioning that as companies move deeper into integration, there seems to be a J-curve relationship between AI adoption and margins. Can you expand on that?
Brandon: Yeah, when we think of the J-curve, we tend to think of private equity, where private equity or a business, makes a series of investments and then eventually it results in value creation towards an end of a time period. And similarly here, if you think about the implementation of artificial intelligence, there appears to be a J-curve that also emerges through a company from an operating margin and revenue perspective. And intuitively, it makes sense because initially if you think about a business, as you're rolling out AI, you might roll it out across the organization and then as you move through, you fine-tune. Maybe you bring in a consultancy to help you fine-tune that model. Maybe you started with a closed-weight model, and you move to your own open-weight model or you try to create your own small language model because you want to have vertical expertise. All that takes investment, both from the ops team, also from your tech team. And I think that process isn't necessarily benefiting revenue yet, and it can be a drag on margins as you're going through that process. What ends up happening though, as you come out the back side of that, is now once it is implemented and you've affirmed your business strategy around it, suddenly then you can then maximize the benefit of that investment and the payoff starts, in our view. And we tend to see that, and I think the study was highlighting that you tend to see that somewhat also in the operating performance across the S&P 500.
Rudy: So how does AI adoption shape the way your team evaluates a company?
Brandon: Well, as we move through the stages and we go from, those initial stages to implementation, we know there's going to be a lot of disruption in the market. One of the things that's been really important to us through this is, trying to understand what businesses have the biggest moats, because I think in many cases, those moats and those walls are going to be bulldozed over the next few years, and so I think every business leader is thinking about the defensibility of their business model and so are we as investors. And so, as we look at the implementation of AI and it moves across, one of the things we're looking for is what businesses are best advantaged to implement, take advantage of it, but then also protect their businesses and their profitability because they have something unique about that business model. It could be a data advantage, it could be a market structure advantage, it could be a product, an R&D advantage. I think this is what makes investing in growth companies so very interesting at this point, because in many cases, you know, there are instances like that across the mid cap space and so, as we transition from maybe looking at more of companies that are directly benefiting AI from the buildout perspective to now expanding our lens to being companies that are implementing and now using it for benefit, we believe, the opportunities really open up.
Rudy: Can you walk us through a company in your portfolio that illustrates that approach?
Brandon: A business that we've really liked has been Natera, and they're involved in the liquid diagnostic space for cancers. Over the last several years, they've expanded from maybe one or two cancer types and now covering a broad range of cancer types for testing for a residual minimal disease post treatment. What's exciting to us though is through that, they've amassed hundreds of thousands of people in a database that they can now use artificial intelligence, and they have recently signed a contract with Nvidia to help them do this, to basically build a product where they actually can help with trial design, help with treatment protocol and recommendation engines, and really use AI in a new world where we can kind of look at the drugs, how they worked on certain types of tumors, and what this particular patient might best respond to. And so, you have natural barriers to entry, in part because of the data that's provided by several years of testing, and allows them to have a unique market opportunity. And so, we're trying to populate the portfolios with a combination of ideas like that as we kind of move forward through this period.
Rudy: Brandon, this has been a great discussion. But before we wrap up, what's one thing that you would want investors to take away from this podcast?
Brandon: The highlight I'd like to leave is we're still early. It's often easy to get caught up in the headlines as the market kind of works through and tries to understand this new technology, price for it appropriately in fixed income markets and data center markets and funding markets, and in equity markets, but the underlying technology continues to move forward. We believe there are very real benefits, and I think as we move from this initial stage of buildout to seeing this in more and more cases in our daily lives, I think it's an exciting time to be an investor.
Rudy: Brandon, thanks so much for being on the podcast.
Brandon: Thank you, Rudy
Rudy: And thank you for listening, for more information on AI or mid caps, please visit alger.com.
