Murray Bender: RBC Investor Services presents insights on the challenges and opportunities facing corporate and institutional investors. Featured on today’s podcast is Nicholas Abe, Co-Founder and Chief Operating Officer of Toronto-based Boosted.ai, discussing artificial intelligence in the asset management space. Welcome, Nick.
Nicholas Abe: Thanks for having me, Murray.
Murray Bender: To start, Nick, tell us a bit about Boosted.ai.
Nicholas Abe: Boosted got its start a few years ago as a structured-data machine-learning company. The idea behind it was, can we help fundamental investors bring AI into their process, effectively taking the universe of stocks that they cover, the methodology that they use to cover stocks, and automate the process of helping them to decide what stocks to buy and sell, have the machine rank the stocks for them effectively, and tell them what stocks the machine’s interested in buying and selling, and exactly why.
More recently, with the kind of advancements in generative AI, we’ve been spending a lot of time also trying to save our users time. And the idea being that while we covered structured data quite well, that we could bring unstructured data in, read the news, read earnings transcripts, read 10-Ks and Qs, and help our users understand kind of holistically what’s going on in the stock market.
Murray Bender: So you mentioned generative AI. What’s a very simple definition of artificial intelligence or AI? And what’s the difference between predictive AI and generative AI?
Nicholas Abe: Yeah. I think that’s a great question. I think there’s kind of a very pedantic definition of artificial intelligence that in some ways doesn’t really matter. And I think when we all think about artificial intelligence, we think of the Terminator. We think of Arnold and making all these calculations about exactly where to shoot his bullets or whatever.
And it’s only been really recently that AI has felt like AI. Right? That whole ChatGPT experience where you get the magic and you type in a question and it gives you kind of exactly what you were looking for or at least something very close to what you were looking for. Predictive AI and generative AI are very similar and, effectively, the same thing. However, the kind of application of the technology is different in generative AI.
So for predictive AI, we can think about things like predicting when a well is going to run out of oil or predicting when you need to change a tire. Right? Use reams and reams of structured data to make these predictions and figure out exactly when something’s going to happen.
In generative AI, if we stick with the text-based elements, what’s actually happening is we kind of throw all of human knowledge into these models and then it learns exactly what different words are used in context with other words and then it tries to predict the next word. So you ask a question, it will look at all the knowledge it has, it’ll look at your question, try and predict the next word. Then once it’s predicted that word, predict the one after that, and so on. And then you get this kind of, again, magical generative AI experience coming out of that.
Murray Bender: So what are some of the ways that AI is currently adding value to asset managers?
Nicholas Abe: From kind of a traditional sense, you’ve always had the top 10 or so hedge funds in the world that have used AI, both on the structured and unstructured side to actually make predictions and actually figure out what stocks to buy and sell. And those managers have always done extremely well.
I think what we’re seeing now is more and more people are trying to bring more and more data into the process. It’s getting harder and harder. We’re asking more and more of asset management employees to try and figure out, how do I use credit card data; how do I use satellite data; how do I keep on top of all the other things that I was doing at the same time.
And so what we’re seeing is this move towards using AI as an automation tool to help automate some of the more—I don’t want to call them manual, but more resource, time-heavy tasks where a machine can go out, do all of the research for you, and then summarize it into something that allows you to decide, do I want to spend more time deep diving into this, or do I kind of get the gist and I can move on to something else.
Murray Bender: So that sounds great. What would you say are some of the challenges that AI poses for asset managers? And how are these challenges being managed?
Nicholas Abe: I’d say there’s probably three primary challenges.
The first challenge is every asset manager out there has their own data that they value highly, and they should, that helps lead them to better decisions. And they don’t really want that data to leave their premises. It’s their IP. They don’t want it to go off, get dumped to ChatGPT, and then have ChatGPT get smarter and then their competitors get smarter as a result of it. So ensuring that your models are operating on your data for your exclusive benefit, I think, is extremely important.
I think the other large challenge is the overall, I guess, privacy of that data. You never want to be involved in some accidental data leak. Right? Where accidentally ChatGPT starts asking questions or answering questions about your customers.
And then probably the third and most important one, I think, is actually institutional inertia. And what I mean by that is it’s really easy to just—these people, like I mentioned, they work 60 to 80 hours a week already. And trying to throw on, hey, how do I figure out how to use AI on top of that job, is a lot to ask. Right? And it creates this environment where it’s very difficult for people to take that leap and say, how do I bring AI into the process or actually spend the money to do it.
And I think that the real risk for a lot of people is not doing it fast enough and then being faced with competitors who have done it a lot faster, who are taking their 60 to 80 hours a week and instead of spending 20 to 30 hours on manual tasks, just using that on higher-level decision-making. Those people are going to win in the long term. The overall costs of trying to implement AI are actually quite low relative to the long-term risks that you face if you don’t.
Murray Bender: So if we look to the future, based on what you’re seeing out there, what does the future hold for AI insofar as the asset management sector is concerned?
Nicholas Abe: Yeah. I think we’ve touched on a lot of the things that I see as kind of the immediate next steps, which is there’s going to be a lot of automation in a lot of tasks where, again, there’s a lot of somewhat repetitive work or somewhat just like research work that AI is going to be able to go out and do that is going to allow people to spend more time on higher-level decision-making processes.
But I think as we get further and further down the line, we’re going to get to the point of hyper-customization for customers. And I guess what that means to me is you’ve already seen cases like Wealthsimple and Robinhood, where basically users can go off and do their own thing. And they interact with it through an app and it’s wildly successful with people who grew up having smartphones.
And I think in the future, instead of offering a mutual fund product or an ETF that is just someone in marketing or someone in product says, this is the ETF that is the future, it’s going to be a user saying, you know what, I really love sports, I don’t like energy, I don’t want to be invested in any defense companies or whatever; go off and make me the best portfolio that you can, knowing what you know about me, knowing what I’m comfortable investing in. And all of that will be automatable, all of that will be able to be achieved at a low cost, and everyone may end up with kind of their own customized solutions.
And I think in that world, people who have the kind of capacity and capability to deliver that customization, will probably end up winning. And we think back about some of the innovations that have occurred in asset management, like ETFs, and I don’t know if there’s anyone that wishes that they got involved with ETFs later. They all wish they got involved sooner, because once people get to their habits, it’s hard to move them off of it.
Murray Bender: So AI is unlikely to go away anytime soon.
Nicholas Abe: I think the real thing for me is when you think about some of the other recent things that have been on the hype train, like cryptocurrency, Web3, metaverse, you struggle to find what the actual use case is. You can’t clearly articulate, well, what does Web3 do for me. Right?
But in AI, we have had conversations, and we continue to have conversations where there’s this really clear and obvious benefit to me to using this technology. And that’s why we’ve seen so much investment in it and why we see so many people talking about it. And I think we’re kind of just getting started here.
Murray Bender: Thanks for your time today, Nick. We really, really appreciate it.
Nicholas Abe: Yeah. Really appreciate it. Thanks for the invite, Murray.
Murray Bender: For insights on a range of topics relevant to corporate and institutional investors, including our previous podcasts, visit rbcis.com/insights. I’m Murray Bender. Thanks for listening.
This content is provided for general information and does not constitute financial, tax, legal, or accounting advice and should not be relied upon in that regard. Neither RBC Investor Services nor its affiliates accepts any liability for loss or damage arising from use of the information in this podcast.