Episode Transcript
[00:00:00] Speaker A: Leveraging AI should not be something that folks should be scared of doing. If anything, it is if it's done and executed correctly. Leveraging AI will give your team superpowers to be able to be more efficient and more productive at the end of the day.
[00:00:16] Speaker B: Welcome to the Institutional Edge, a weekly podcast in partnership with Pensions Investments. I'm your host, Angelo Calvello. In each 30 minute episode, I interview asset owners, the investment professionals deploying capital, who share insights on carefully curated topics. Occasionally we feature brilliant minds from outside of our industry, driving the conversation forward. No fluff, no vendor pitches, no disguise marketing. Our goal is to challenge conventional thinking, elevate the conversation and help you make smarter investment decisions, but always with a little edginess along the way.
Today's topic is AI's winner take all.
How will AI reshape the institutional investment industry?
And my guest, AJ Hashim, Vice president and co lead of Berkshire Global Advisors Financial Technology Practice, is the perfect guest to discuss this topic as he sits at the industry nexus of growth and innovation. Aj, it's great to have you on the show today.
[00:01:16] Speaker A: Thanks, Angelo. It's a pleasure.
[00:01:18] Speaker B: We've had these conversations, aj, over the past five years, six years, and really today this is just a continuation and maybe it's a chance for us to pull together a lot of the thoughts you and I have had. I've always found when I'm doing my writing and my thinking about AI, I somehow eventually come to you as a reader and say, aj, help me think about this problem. And I think I do that, aj, because you have such a kind of an outsider's perspective. You're neither an allocator nor a manager, but you kind of sit, you know, like in that nexus in the asset management business, you see a lot, you hear a lot. And that's why I think this is a kind of a, it's a good topic for our listeners to move a little bit outside that allocator sphere and get kind of a perspective from like a real industry professional that looks at both.
[00:02:09] Speaker A: I definitely enjoy all of our conversations. It's fairly easy to just kind of get lost on the phone with you and look at the clock and all of a sudden it's two hours later and we're still talking or going down the rabbit hole on whatever topic you feel like discussing.
[00:02:22] Speaker B: But hey man, this is gon 30 minutes plus or minus. So we gotta, we gotta pack it in here.
[00:02:27] Speaker A: So exactly.
[00:02:28] Speaker B: Let's jump in and ag I know from our conversations about, you know, how AI is transforming the asset Management business. Your view has always been, not always, excuse me, it's recently been AI is table stakes for asset managers. Why is that the case?
[00:02:47] Speaker A: You know, just kind of seeing how AI has impacted every single other industry, you know, across all verticals, whether it's health care, financial services or more broadly, tech. You know, I think it's inevitable that, you know, asset managers are going to have to leverage artificial intelligence within every aspect of its business, whether it's from the idea generation, workflow, through trade execution, and even back and middle office functions. If you do not apply AI to your day to day business, quite frankly, it's either going to be extremely expensive to run your business or you're just going to be left in the dust.
[00:03:24] Speaker B: Is that what you're hearing from Berkshire's clients? I mean, they're looking to adopt, integrate AI into their businesses and perhaps on the other side, those strategic targets that are coming to you saying we're looking for partnerships, is it really is AI front and center right now I can.
[00:03:43] Speaker A: Speak to the fintech clients because that's my main coverage. And the answer is yes for sure. A lot of these companies are looking for ways to streamline their operations.
Previously there was a massive shift to moving to offshore operations for cost effectiveness. However, what we realized was that the quality of work or even the time that it takes to get things done, just simple due to humans being in the loop, presented a challenge. Now with the implementation of AI, this can be done on a 247 basis, you know, fairly efficiently in a far more cost effective manner than leveraging humans. And I will say, because it's simple to kind of, or it's easy to go down the rabbit hole of will AI take over jobs? You know, I think my, my personal view and the view of many others that are in the ecosystem is that leveraging AI should not be something that folks are, should be scared of doing. If it's done and executed correctly, leveraging AI will give your team superpowers to be able to be more efficient and more productive at the end of the day.
[00:04:50] Speaker B: And those superpowers, that's now an example of table stakes in your mind, you've got to be there. You can't be waiting two years to get there. You've also mentioned in the past, AJ the other benefit is making better investment decisions.
AI should allow asset managers to make better investment decisions. Is that also in the realm of table stakes?
[00:05:13] Speaker A: Yeah. So I mean, as you know from your asset management days, it's all about being able to access the information whether it's quicker or be able to read it better than your competitors. And that's, that's where the true alpha is. Right. So with the application of AI within asset management, you know, we are entering a, a new age where folks are going to be able to analyze data in ways that historically would have been extremely expensive. It would have taken a long time to do, and quite frankly would require a skill set that many firms perhaps don't have. Right. So being able to look at things, whether it's alternative data, private data, public data, and basically pull it all together and throw a contextual analytics layer on top of that to be able to access new ideas that can ultimately generate alpha, I think is where the industry said it.
[00:06:11] Speaker B: So if that's where we're headed, how do people get there? I mean, you know, you're an investment banker, you're involved in transactions that I assume are helping people get there. In our conversations, I'll just say quickly, you know, we've talked about asset managers build by lease. That's the approach they could take to gaining these table stakes and gaining a competitive advantage. So how do you get there? I mean, as an investment banker, what's your view?
[00:06:37] Speaker A: Well, more broadly to your point, there's three options, right? Buy, build or lease. I would say the leaders in the quantitative asset management space are most likely going to build that internally. They have an army of data scientists that are able to create these proprietary systems and at the end of the day it's about a competitive edge. So I don't think they're going to necessarily use something that their competitors are using, which then leads to are you going to buy the system or are you going to lease it? Now you have serial aggregators that are consistently looking for new technologies, new teams that they can basically roll up into their existing enterprise and basically continue to build internally. Now you can think the blackrocks of the world and some of these other serial acquirers and then you have, where I think most of the, let's call it mid size to small size asset managers, where they may not necessarily have a blank check to be able to go out there and pursue a roll up strategy, but they want to stay competitive by being able to leverage tools that are out there that can help them stay competitive.
[00:07:45] Speaker B: Okay, so in that case the smaller to mid size are more likely to do the leasing, which in my view is they're contracting with a third party, I'll use a simple word, software provider.
[00:07:56] Speaker A: Correct, correct.
[00:07:57] Speaker B: Aren't there certain risks in that though from using a third Party.
[00:08:01] Speaker A: Yeah. So I think it depends on which third party you're using. I mean, this, this is more of a legal question. But obviously access to your proprietary data is a huge risk. Hallucinations are a huge risk. So there's a lot of nuances that, you know, one needs to consider before, you know, plugging into some sort of, you know, existing third party solution. However, from my understanding, a lot of these guys understand that and they have been able to create solutions that can address those concerns, such as, you know, whether it's a private cloud or it's an on prem solution. So you don't run the risk of, you know, cross contaminating your data with public data and then vice versa. That's the last thing you want to do is have somebody else use a model that was trained on your proprietary data.
[00:08:45] Speaker B: And there's also the supply chain risk. The company you use goes out of business. But.
[00:08:50] Speaker A: Right.
[00:08:50] Speaker B: You know, there's two things that come to mind. The first is you're talking about larger established firms, quants, they seem to have the advantage of building something proprietary.
[00:09:03] Speaker A: Right.
[00:09:03] Speaker B: And then you're talking about the third party leasing arrangements with smaller to mid size.
Wouldn't that give the larger managers a competitive advantage because they're building this in house? Because they have at least a checkbook to do it, doesn't mean they'll be successful. But if we assume that, it seems like there's a schism developing in the industry.
[00:09:23] Speaker A: Right. Yeah. And I think I agree with that. It's always been a game of creating a competitive edge. And again, it's whether it's the access to the right information. Right.
And you get the information through data. So now some of these larger asset managers most likely have a bigger budget and access to whether it's proprietary data again, or, you know, public and private data or even alternative data. Right. So being able to ingest that and look at it in ways that, you know, your competitors can't is definitely an advantage. And in some cases you'll see that, you know, some of these larger asset managers might acquire some of these alternative data providers because they don't want that data out there to be used against them.
[00:10:04] Speaker B: And that itself creates a competitive moat. If you have proprietary data, that's relevant to the research question.
[00:10:11] Speaker A: Correct.
[00:10:12] Speaker B: You know, you shut out your competitors.
[00:10:13] Speaker A: You may not shut them out, but you might make it a little bit more of a headache for them. Right. Because the data's out there. Right. It's just doing the work of curating it and managing it in A way that's accessible.
[00:10:24] Speaker B: So it also kind of brings up in my mind, if you have this schism, is it going to lead to some kind of consolidation with the smaller to mid sized managers?
Because I was a small AI manager, as you well know, we did a great job. We built some amazing products. We lacked distribution was our challenge. And distribution in the sense of having people understand what the heck reinforcement learning is. But we kind of got squeezed out of the industry just kind of through attrition, not through bad performance or any internal problems. But we're the winners and losers in this. It seems like it's going to be the larger firms that have data, budget and other attributes.
[00:11:09] Speaker A: Yeah. So from a consolidation perspective, there's really two main plays. Right. So it's one, whether it's a tech play where you're either acquiring the data or the systems that some of these smaller managers have developed and then rolling it into your distribution channels to be able to basically scale and grow is the first strategy number two. I think it's a talent war. Right. So everybody is out there trying to hire data scientists and get them to create some of these systems internally. So it wouldn't surprise me if down the road that there was a massive consolidation of some of these smaller managers that have the brain pool to be able to go out there and create some of these systems.
[00:11:50] Speaker B: What are the key attributes that you envision the winners in our industry will have? And we're not talking about alpha and being able to generate alpha. That's a given. You have to be able to do that. But think about it from your seat. When you're looking at these companies, you look for certain attributes. If you're trying to connect the large asset manager with a talented smaller asset manager. So what are they?
[00:12:18] Speaker A: I think it ultimately comes down to three things. The first one being the culture at the firm. The true winners are going to be the ones where from the top to the bottom of the organization, understand the opportunity here. With AI, it's not a uphill battle consistently trying to fight the board on rolling out a new AI system. It's at the board level. They understand that this is the future and that they are creating an environment internally that promotes the use of artificial intelligence and are consistently looking at ways to grow by leveraging it.
[00:12:50] Speaker B: AJ Let me stop you there and ask, when you're dealing with your client base, are you typically dealing with the C suite? When people are thinking about technology? They are. Okay, yeah.
[00:13:01] Speaker A: Right.
[00:13:01] Speaker B: So already in your conversations it's coming from the top down.
[00:13:05] Speaker A: Yes, in most cases. I mean there's always going to be in certain cases where folks are skeptical about the, the technology and whether it's, they, they understand it, you know, that's because they understand it or not. You know, that's a different, that's a different case. But in most cases I think within the technology, the fintech space, most people understand, you know, the use case here.
[00:13:25] Speaker B: Okay, so go back. You said there were three attributes, one being, you know, culture, C suite, conviction, et cetera. What are the other two?
[00:13:34] Speaker A: Number two is, and I kind of alluded to this earlier, is the access to the data.
Now there is, as you know from working with AI models and even most people like it's same thing in Excel, right? It's all about the inputs that you, that you plug into your model. So if you're plugging in low quality non proprietary data, you're just going to have average, if not worse outcomes. So it now becomes a game of who's got access to the highest quality and most proprietary data to be able to plug into these systems. Which leads to the third part of that contextual layer and it really being able to look at the data that's presented to you in ways that historically was difficult to do. So whoever can do all three of those together I think is well poised to succeed in the future.
[00:14:25] Speaker B: In some of your earlier comments today, AJ the subtext is also talent. I mean you mentioned you could have data, you could have kind of the right culture, you got to have the right people. And I guess that's the one of the attributes of the third feature you mentioned, being able to create the AI, you know, getting that talent. So when you're working with prospective clients, I mean, do you kind of have a checklist to see at the first level, the cultural level, that there's going to be a fit because you can see there might be a smaller manager or maybe it's a service provider that's doing some really cool stuff with LLMs and now a larger firm wants to buy it. Do you assess that cultural level or layer?
[00:15:10] Speaker A: Yeah, so I think that's one of the most important things. So when we're looking to onboard new clients, I think one of the most important things is looking at the management team. We typically work with managers that have a true vision, that have done a great job of building the business to where it is today and now they're looking for the next chapter of growth. I think the second piece is the underlying technology. There's a lot of companies out there that claim they're doing something that is innovative. However, you know, I've seen plenty of other companies that are doing something similar. It may not be 100% the same, but in the eyes of a buyer it might as well be the same thing. Right? So looking for companies that are offering a true differentiated service or solution is the next one. And then the last thing is really the growth opportunity of this particular business. Right? So basically looking at a Rolodex of relationships and seeing what is interesting to a lot of these folks. You know, we have consistent a regular dialogue with the corporate development teams at all of these top firms and they're consistently telling us, you know, what is interesting in their space. In some cases they want to look at everything, but in most cases, you know, they'll be fairly specific about what mandate that they're looking to execute. So as a banker, it's my job to go out there and go out there and find great, great companies that one again have great management teams that have great products and that would be a great fit for ultimately these buyers.
[00:16:39] Speaker B: It's interesting, you talk about how you interact with your clients, the corporate development team, and they say what they're looking for.
One topic you and I have discussed, and maybe this is a good segue to it, is the whole idea of agents of multigenic systems. You and I offline, have talked about this, I've written something, you've reviewed it before I publish it. Are any of your clients talking about these multigenic systems or using agents?
[00:17:07] Speaker A: Again, it's hard for me to talk about the asset management side, but on the fintech side for sure, especially some of these startups, it's actually quite interesting the way these systems work, right? So you have like the worker bees, quote unquote, then you have the supervisory bees, and then you also you have like the group heads. So it's basically an agentic enterprise, so to speak, that is fully autonomous and it does the self checking and at the end of the day it just produces the deliverable that ultimately will be reviewed by a human and then reiterated based off of the feedback that the reviewer ultimately provides. But yes, I mean the answer is yes. I think agentic systems is the next wave of AI adoption. But quite frankly, I think right now a lot of organizations are still stuck in that data cleanup, data management process of just making sure that their underlying data is ready to be plugged into some of these more complex systems.
[00:18:03] Speaker B: You know, you have clients that are looking for help solving that, not the agency system, but cleaning up the data because we're still then back at table stakes to be honest.
[00:18:13] Speaker A: Right.
[00:18:13] Speaker B: Is that really what people are looking for help on? Your clients are looking for solutions there?
[00:18:18] Speaker A: Yeah, I mean, they're not necessarily coming to us as investment makers to help them clean that up. If anything, they're either doing it in house or there are other third party service providers that are brave enough to take that challenge on. But you know, I think that's where, where most people are today.
[00:18:32] Speaker B: It's kind of interesting because when you and I were working together, you know, with Julia and Bruce and the team, well, we certainly weren't using an agenic system at the time, but we were using a self learning autonomous system and you know, we struggled to have people understand what that even means.
It wasn't so much the human in the loop issue, it was the idea of self learning systems. Are you finding, you know, now people are getting more comfortable with this idea of self learning systems?
[00:19:05] Speaker A: Well, to an extent, you know, ultimately I think we're not at the point yet where folks are confident with the fully autonomous decision making when it comes to asset management as you know, with like trade execution and whatnot. But the low hanging fruit like you know, back, back office, middle office optimization, you know, I think people are more open to that ultimately before a trade is placed or you know, serious money is moved around, you know, that I think that that will definitely require some sort of visibility at least into how the decision was made and then ultimately the portfolio manager or whoever it is that is ultimately responsible for the, for the decision will make that decision.
[00:19:48] Speaker B: Well, that kind of causes me to think that multi hygienic systems may not be coming anytime soon because they tend to be rather opaque in terms of how decisions are made.
[00:19:59] Speaker A: It's hard to say that because as you know, the industry is consistently evolving and there's always some new thing that's popping up every day. But as of right now, it seems like we are a ways away from that.
[00:20:10] Speaker B: Right. I'm going to go big picture on you here. I'm going to move away from table stakes, I'm going to move away from those. You talked about the features, you know, the attributes of the prospective winners. Asset management is a very lucrative business and you've got hyperscalers like Google and Microsoft and Amazon.
They have all the attributes of a future winner. I mean, you told me it's culture and conviction. You need to have the compute, the data, the talent, the budget, the AI. Do you see these firms kind of moving in to the asset management business and maybe Building or buying or partnering on the asset management side.
[00:20:49] Speaker A: Yeah, no, I mean, look, I think these guys have strategically positioned themselves as technology companies. I think, you know, you know, the one word that comes to mind is regulation. Right. So being a regulated financial services player comes with a broad list of headaches that are necessarily that are associated with that. Right. So, you know, I mean, you have, you know, Amazon with their Amazon pay solution, but I think there's back end, I could be wrong on this, but I think they have back end partners that actually do the banking side of things, whereas they're just the technology provider on there. It's more likely that these players are going to partner with firms, whether it's through a strategic partnership or a strategic investment. Like for example, Google ventures led a $30 million cap raise into Wealth.com recently, an AI powered estate planning tool. Right. So I think that's the direction that would probably makes more sense for those players.
[00:21:45] Speaker B: You know, when I was starting Rosetta, I went to Google X and I said, let's partner. I mean, I told them this is a diversifying revenue stream. If you could get in it, the margins are good and you've got all the attributes. And they're like, no, it's not something we're going to do. I thought they were doing moonshots. But no, let's wrap this up. I mean, this has been great. I want to go back and touch on something that you mentioned at the very beginning of your remarks. You talked about how AI is table stakes not just for asset management businesses and fintech companies, but also for basically all verticals, all businesses. What the heck is Berkshire doing in the AI space? I mean, you guys are a leader in the M and A, you're big on fintech. Are you doing anything?
[00:22:30] Speaker A: So we are analyzing ways internally to basically streamline some of these workflows. At the end of the day, if I can free up an extra 10 hours a week or 20 hours a week for our analysts, you know, that would mean a lot to them. So there are a lot of basically tedious tasks. One example, you know, we're building a proprietary market intelligence platform that, you know, we can use to track transactions. And you know, right now that's a fairly manual process that is basically done by the bullpen. Right? So being able to use NLP and you know, plugging into various data sources, whether it's public data, private data or CRM or even alternative data, if that makes sense, in a particular scenari, and organizing this in an automated way so I can Save my team some time and to allocate to other more impactful tasks is the low hanging fruit right now. I think we're just like many organizations. We are in that process of cleaning up our internal data to make sure that it's ready for this next wave of AI solutions.
[00:23:34] Speaker B: So is it a build, buy or lease decision?
[00:23:37] Speaker A: It's a combination of leasing and building.
We are working with a few developers that have platforms that allow us to build custom applications on top of them and through their teams they're able to create these bespoke applications that we can basically leverage. And this is just one application that we're looking to roll out. There are a bunch of other use cases that we're looking to from every function of whether it's HR compliance or even the front office function that could be automated through AI.
[00:24:11] Speaker B: Why don't you just hire me and I'll build your multi agency system. I'll solve the whole problem for you.
[00:24:15] Speaker A: Send the resume, we'll get you resume baby.
[00:24:18] Speaker B: I'm too old to be working for anybody. This is all about me as a consultant, you know, just like you guys are consulted me as a consultant. But no.
[00:24:26] Speaker A: Well, we'll have to. We'll have to get Bruce.
We'll get Bruce on board.
[00:24:31] Speaker B: And you mentioned Bruce. Bruce Cameron. I know from our discussions you are definitely hitting the C suite level with this, that you guys are at Berkshire are committed, you know, from top down to do this. So it kind of fits with your attributes. So.
[00:24:46] Speaker A: Right.
[00:24:46] Speaker B: But this is cool. AJ this is great. Thank you very much. I appreciate it. I always learn a little something from you. Not much, but a little something.
And I appreciate you playing along.
[00:24:56] Speaker A: So thank you so much Angela. I appreciate the time.
[00:24:59] Speaker B: Thanks for listening. Be sure to visit PNI's website for outstanding content and to hear previous episodes of the show. You can also find us on p and I's YouTube channel. Links are in the show notes. If you have any questions or comments on the episode or have suggestions for future topics and guests, we'd love to hear from you. My contact information is also in the show notes and if you haven't already done so, we'd really appreciate an honest review on itunes. These reviews help us make sure we're delivering the content you need to be successful. To hear more insightful interviews with allocators, be sure to subscribe to the show on the podcast app of your choice. Finally, special thanks to the Northrup Family for providing us with music from the Super Trio. We'll see you next time.
[00:25:50] Speaker A: Namaste the information presented in this podcast is for educational and informational purposes only. The host, guest, and their affiliated organizations are not providing investment, legal, tax, or financial advice. All opinions expressed by the host and guest are solely their own and should not be construed as investment recommendations or advice. Investment strategies discussed may not be suitable for all investors as individual circumstances vary.
