Andrew Brzezinski: Separating AI Hype From Reality

This podcast is brought to you by Wells Fargo Advisors. Looking for a firm your clients can't outgrow? Wells Fargo Advisors offers financial advisors full balance sheet solutions with wealth planning, investment products, banking, and lending that meet the needs of affluent to ultra-high net worth clients. Build your practice your way and on your schedule with independent and employee model options designed to meet your needs today and tomorrow. Learn more at joinwfadvisors.com.

There's a lot of hype around AI since the release of ChatGPT 3.5 back in November 2022. What's real? What's hype? And what should you be doing with it right now? Hi everyone, I'm business coach Steve Sandusky for Barron's Advisor, The Way Forward podcast. My guest today is Andrew Brzezinski. Andrew is head of data strategy Business Intelligence and Advanced Analytics at Fidelity Investments. In today's conversation, we explore the practical applications of generative AI, the barriers to adopting AI and how to overcome them, the key risks and ethical considerations of AI, and the future outlook for AI in an advisory practice. With that, please enjoy my conversation with Andrew Brzezinski. ChatGPT was released, or at least the version that entered the public consciousness was released about a year ago. Obviously, there's been a tremendous amount of hype around artificial intelligence since that release, but AI has been around for a long time. I'd love for you to set the stage a little bit here with where are we today? Help us separate the hype from the reality.

Over all these years, it's generated a ton of excitement about the possibilities of what it can do. And the hype part, of course, is that people are getting really excited and they start overestimating the potential of AI, and they get to thinking that it can solve virtually any problem. So with ChatGPT, that's an example of a generative AI, or what is more technically described as a large language model approach to AI. And so ChatGPT is basically the most famous example of that in the world today. You have to try it out in order to get a feeling for what it can even do, right? It's really remarkable technology. It basically allows you to ask almost any kind of question, and it usually gives you a pretty good response. And that's irrespective of the subject area that you're asking your question about. And moreover, the thing about ChatGPT especially is that it gives these responses that sound chatty and human-like. So it gives you the feeling like all the knowledge of the world is at your fingertips. So the hype part about ChatGPT, and I guess other AI technology, is that it— people have this impression that it can solve anything. But you can ask it simple examples that show that it can't do everything. So throw a question into ChatGPT that says, what's the weather today? And it's going to tell you, I can't answer that because I don't have real-time information. And it has to do with the fact that this algorithm was trained with a certain dataset, right? For most people, you can ask it a question about, tell me something personal about myself. And if you're not a public figure, it might not be able to tell you anything about you, right? Again, there's a lot that ChatGPT can do fabulously, but there's still a lot that it doesn't know and therefore it can't do. The one other thing I should mention about hype is around like the disruptive potential of AI technology, right? It's true that applying AI in the working world is going to change the nature of tasks and create efficiencies. And that's been the story of technology forever. So while technology and AI are becoming more a part of our lives and the way we work, we've always been really effective at finding ways to work with that technology. And so it's a combination of people and computers that continues to be the most realistic outcome. In the end, the way I look at it, with the entire AI journey that's been going on for years, and even this latest iteration of ChatGPT. It's about pragmatically looking for new and impactful ways to use data to solve problems for your business, your firm, your clients.

In the financial advisor world here, it seems like one of the most prominent applications of ChatGPT and similar AIs is in marketing, is in content creation, is in generating emails, and just really more on the whole marketing side. So that I think is definitely a great practical application. And you mentioned the research. ChatGPT is great at asking it questions to get background information on stuff that's been around for a long time. So I personally have found lots of uses on the marketing and the content creation side. I'd love for you to share some other practical examples that you're seeing from your perch there at Fidelity in terms of how firms are using it, how technology companies are starting to incorporate it. Because again, I think there is all this hype and I'm not sure that I'm seeing a lot of really concrete, practical examples that are really moving the needle for advisors when it comes to like operational efficiency, providing better customer service. I'd love to get your take there.

Yeah. And Steve, I, I think the way we have to look at it is that although it's been a year, although there's been a long AI journey, this is really early days. And so with these examples that you're providing, like we're seeing like the dabbling going on across industries to say, yes, there is something that's transformational here. Yes, we've never seen an AI capability being able to answer questions virtually about anything in a pretty good way and in a chatty human-like way. So let's try to find ways of applying that. I completely agree with you that the best use cases right now for generative AI have to do with summarizing documents, right? And maybe that means taking something very technical and translating it into simpler talking points. Summarizing something technical also into an email. You could start with a white paper and then you could say, what would be the email that introduces this white paper? What would be the social media post or blurb that we could put out there that helps to catch people's eyes and attract them to the technical document? I think that those are legitimately many of the use cases that are being played with this technology right now. And, you know, soon enough, the hope is as we work through those, that we can find ways of leading them to more impactful opportunity areas, as you described, leading to operational efficiency or service excellence or personalization of some kind. But again, we have to recognize that it's still early, right? And industries are all scrambling to try to adopt this at this point.

The idea of big data has also been around for a long time, and I think there's some potential promise here that with AI we can do a better job of extracting insights from all this data that we've been collecting here for years and decades. What is your current thinking in terms of how AI can be used to glean insights from all this data? Again, getting back to a practical level, from a financial advisor. It's early, I get that, but what can you see on the horizon? What are you guys working on that might enable us to use AI to get great insights from big data?

Data is the key. I'm glad that you focused on data with this question because no AI technology, no AI solution, no analytics are going to be useful unless they have good data. And so the way that we're trying to look at this is pragmatically, what is the data that's available to AI technology and what questions can then it help to answer, right? So the way I imagine this doesn't necessarily have to be a generative AI use case, but we could be talking about other things like predictive analytics or anomaly detection or recommendation solutions, those types of things. That's the broader ecosystem or set of capabilities that exist around here. So one of the things that I am especially interested in and that we're working on is to try to identify atypical events in the data, right? Where it may be simple things like observing transactions or patterns around behaviors or holdings and positions and those kinds of things, like nuts and bolts type of data that you can then say, what am I observing in there? Are there anything that, that seems to be anomalous or atypical for an individual client versus others. So then there might be some sense of benchmarking or segmentation there. And so you can come up with a series of interesting insights, some of which are not necessarily earth-shattering. But then what you can do is to say, I have all of these insights now. Let me start thinking about which one I can recommend as the best thing for Steve to consume today. You want to get to something actionable. And so what is the right thing for you to engage with right now? Who do you call? What do you talk to them about so that you can have the greatest impact for your book of business, for your relationship, that type of question.

This idea of a fully integrated suite of tools that an advisor uses where you just enter the data in once and you don't have to enter it a second or third time into different systems has been the Holy Grail forever, and we haven't reached it yet. If you go to the warehouses, for the most part they have that because it's one self-contained system, and they may not have the best CRM system, they may not have the best portfolio accounting system, but it works. It all talks together. And that's one of the benefits of being with a warehouse is you get that. Now on the independent side, when we throw AI into this and we throw big data into this, is that going to massively complicate the ability to create an integrated system and take Fidelity and the other custodians as an example? You guys have the data. So are you guys in the catbird seat because now you've got the data and everyone else has to come to you for the data? Or how do you think about whether AI is going to make it even more difficult for firms to get this fully integrated suite of services that all talk to each other?

You're homing in right on the greatest challenge in this industry. Right? So the fact that there's so much technology choice is, I guess we have to put it as like it's both a blessing and a curse, right? So firms can choose their set of tools that are appropriate to their specific needs and the way that they want to manage their businesses. That's wonderful for the specific tasks and what those tools are going to be trying to accomplish for them. When you try to say, but then on top of this, I need to accomplish a big data outcome and I need to put AI on top of that solution. It's not that AI complicates things. It's just engaging into a complex complex situation, right? Of course, then it's challenged because there are so many variety of tools with inconsistent data models or approaches to the data. The data doesn't naturally talk with each other. And so therefore, it becomes incumbent on firms who often, that's not their top of mind thing to do, to try to pull together a data strategy for themselves that then they could put analytics and AI on top of that. So like you say, larger players who have more access to a wider swath of data are in better position to help out in that case, right? Custodians have access to a tremendous amount of brokerage data, but not necessarily always the full book of business, not necessarily always to all the planning information, all the suitability information, CRM type of data. And so even if Fidelity is challenged, although we have a pretty remarkable and appealing data situation, No one truly has all the data except firms, and then in many instances, they have a challenge of pulling it together. So I guess in summary, I have to say it's not perfect for anyone. But if you take a look at what technology like ChatGPT is doing, it's cutting past some of that. And it's still early days. I'll keep saying that in this conversation, but there's something remarkable about the fact that it could index across like a variety of inconsistent documents and knowledge repositories across the internet and synthesize synthesize that into a single unified dataset that it trained for itself. So I think we can map ahead to a future in this industry where similar things are happening, that AI technology can bridge the gap across a complicated technology ecosystem and start stitching together data that isn't easily able to talk with one another and then giving you insights on top of that data overall. Again, I feel like that's probably a years-in-the-making type of vision, but it is out there. This is, I think, a lot of what people are seeing as they see the generative AI transformation happen?

When it comes to technology, I think advisors are in a continual state of frustration because we're always in a continual state of becoming when it relates to technology. And what I mean by that is it's always changing. Like, just when we thought we had it nailed down, all of a sudden ChatGPT comes out and we've got all this really cool generative AI. We've got Midjourney and we can create amazing images and all these different things. So it's, there's there's always a new thing coming out and we never get there. We never can put a stake in the ground and say, I've got my tech stack, this is going to work for the next 5 years. It's not because it just changes so fast. And we all are subject to the shiny object syndrome where we all want to get the latest and greatest thing. And so it's— I think we're always going to be frustrated. Now I'm taking the devil's view here, so to speak, and I'd love for you to push back on me if you can and say, no, Steve, you got it wrong. We're going to get there. It's going to be cool.

Everything's going to work.

Just hang in there. Am I wrong? Tell me if I'm wrong here.

It's all a matter of perspective, Steve. So it's really easy to look ahead and just to say, oh my goodness, look at all of the wood we have to chop. But you can also take a look in the rearview mirror and say, look how much technology is part of our lives. Look how much technology has transformed the way that we work today. It hasn't been perfect. It feels cobbled together. And yet it's deeply part of our lives and the way we function. So if you zoom ahead a year, 3 years, 5 years, 10 years from now, it'll be the same story. There will be the new shiny objects, the major opportunity areas that the next iteration of technology and AI and whatever is bringing to us. And yet we'll be able to look back and say we made remarkable progress in that time. We know so much more. The data is so much more connected. The tech feels that much more integrated, right? Where these pain points exist, it's an opportunity for fintechs, companies, others to try to help solve problems. And we're seeing that innovation happen all the time.

We certainly have come a long way. And I'm old enough that when I was in college, and I took business data processing classes. So I would be in one building on a terminal and I would do my coding in Fortran and COBOL, and then I'd have to go across the street to another building and wait 10, 15, 20, 30 minutes to get my output to see if my program worked. And this was on punch cards. So I'd have a stack of punch cards and if the program didn't work, I'd have to thumb through the punch cards and go to the key punch machine and punch a new code in and insert that in the deck. And it's, yeah, we've come a long way certainly since those days. But it also highlights the point that we have to be continuous learners. And while we may have this love-hate relationship with the technology, we absolutely cannot avoid it. And we have to stay on top of this. It's not that we have to be on the bleeding edge of the technology or the leading edge of the technology, but we've got to be on the technology. We've got to be with it. Let other people be the trailblazers. That's totally fine. But we can't ignore it. We got to stay on top of it. So I think that's again another, maybe frustration's not the right word, but we have to be continuous learners and adopters when it comes to this technology.

Yeah, it's really well said, Steve. Like my sense of this is that it ends up being a necessary ongoing investment within our firms and our associates and our people. Like the change will always be there. Not adopting it is very risky. Because people are finding clever and creative ways to adopt data, adopt AI tools, find efficiencies in the way that they work. And so it's sometimes only in subtle ways, but they're accomplishing more personalized relationships, more efficient outcomes, ways to scale their book of business and their practice. And those kinds of things like add up over time and it could leave a firm, an advisor behind if they don't adopt.

We definitely run the risk that with all of this technology, with social media, with our faces in screens pretty much all day long, that we can just get so caught up in not having the human-to-human connection. And we're going to get lost in this data. We're going to try and be analyzing the big data. We're going to be using ChatGPT and all that good stuff. So it seems to me that while on the one hand, we need to go deep into the technology. At the same time, we have to go the opposite direction and we have to go deep into the human one-to-one connection. And it's the human one-to-one connection that can never be scaled. So we talk so often in our business about, hey, I'm trying to scale my business. Yeah, you can use technology to do some of that, but at the end of the day, there's no technology that's gonna enable you to speed up the one-on-one relationship that you have with the client. So I'd love for you to share any other thoughts that you have in terms of how can we use technology to personalize the relationship, to enhance the relationship, to support the one-on-one client communication? Are there some applications out there that are gonna enable us to deepen those relationships?

Yeah, it's really well said that there is a human element that should never be lost in all of this. And I think ultimately the way that we should look at technology in general and the AI that's happening all around us is that it's an enabler of more time spent on the human aspects of the work, right? And so for the financial advisor, I'm thinking of those really deep deep and intimate relationships, that feeling of trust, the ability to really feel that you're looking somebody in the eye and they understand you. That's stuff that the AI is, is not going to be able to replicate because you're dealing with a person who's had a lifetime of experience. AI will just never be able to compete there. So that's where the advisor should play, right? And the firm at scale is going to have a lot of advisors who do that type of work. But you're right, it's not like suddenly you could say this affords me the opportunity to do that at 10x the scale that have had up until now. What kind of personalization is AI helping with? Again, it comes to the data, right? Increasingly via the systems that we're using today, we're amassing a greater set of information about investors, about clients, right? And so that more holistic understanding of the client gives you an opportunity to engage with them more proactively and on a more holistic personalized basis. One of the simplest ways of personalizing, of course, we call that AI is just to say, here's a proactive reminder that it's a birthday, call the client today, that type of thing. But you can get into more sophisticated use cases that actually require more advanced analytics and AI and prediction, which might be around segmenting a book of business, understanding the client and how to treat them and how to treat clients like them. And we hear about those kinds of AI use cases, like people like you buy X or want to be treated in a certain way, that type of thing. So having an advisor kind of already loaded up with some of those preferences and personalization opportunities and suggestions about here's who you wanna reach out to today, here's the topic that is gonna really resonate for them. Here might be some kind of talking points for that conversation based on conversations in the past. So you seem like you have a follow-through based on the market conditions that exist today, the economic situation, their investment preferences. You bring bring all of that more holistically together and you can end up with a hyper-personalized, deeply relationship-oriented conversation. The advisor delivers that, but they might have a technology that kind of cues them up and makes sure that they're ready proactively for the conversation rather than something that often feels very cookie cutter when you go in for the advisor conversation. Investors have higher expectations. They want to be known. They want to be engaged in a way that feels like they're understood. And so this technology is going to put advisors in a position to come across that way.

If we go back 100 years or so, there was a thought that with advancing technology, there's going to come a point where we're going to have a new leisure class because all this technology is going to be doing all this stuff for us. Now, of course, that didn't happen. And it seems as humans, instead of taking back some leisure time, we just do more stuff and we still work and we still get stressed out. Do you ever see that changing, or do you think that with all of these advancements in technology, it's just going to continue, that we're going to be able to do more for our clients and we're not really going to get back our time and have more leisure time because we're just Type A personalities and we just want to do more stuff? Is that what we're stuck with as human beings?

Steve, you're getting into the nature of people and American culture and all sorts of things, right? We like to work hard. And so if we find tools that give us an opportunity to like more effectively do our jobs, many people will choose to still have a very full day or a very full week. But I have to tell you that I've had instances where we're prototyping tools or working with analytics where we're showing an insight to a client and their reaction is, this is really great. That'll save me some time and I'll either get to spend more time on more deep and richer conversations, or I might be able to spend more time with my kids. Like we've literally heard those kind of feedback. Now the question is just what is the longer term outcome? Everybody individually is gonna make their choices. And I imagine that still you zoom forward 50, 100 100 years, like the way that you, you look back, and probably we'll all be pretty busy. That's just the nature of work that we have around here.

I want to go back to something I said here a few moments ago about you don't have to be on the bleeding edge of technology or perhaps even the leading edge. I'd love to get your take on that, meaning, does an advisory firm have an ability to gain a sustainable competitive advantage if they are an early adopter of some of this AI? If they go headfirst into technology, is that a sustainable competitive advantage or is it not because it just changes so fast and there's always someone that's gonna be nipping at your heels?

Yeah, being right at the head of the pack is costly because you're dabbling with technology that may or may not play out. It may or may not be mature. And so it takes a certain fortitude to do that. But now you could go back into that, that front part of the pack. There's an interesting opportunity there. And again, it's not necessarily for everyone, but firms who would want to consider this should be of the mindset that I'm doing something that's a more holistic benefit to my business, right? It's not necessarily about creating and delivering the AI insight, but there's a more end-to-end thing that has to happen to effectively adopt AI in the workplace. For example, As you work through use cases and you create, say, recommendations for an advisor or an ops person to take up and act upon on a daily basis, what does that mean? You have to have assembled all of the data into a place for the analysis to happen. You have to put the documents into a central place. So there's just like the good housekeeping of assembling a data strategy, getting it normalized, getting it cleaned up and ready for analytics. Analytics and AI. Inevitably, most of us will have to go through that exercise. I suppose we can wait a long time for some vendor or provider to do it for us. But again, those who are leading in the pack, they'll be able to do that at some expense and get ahead of the data strategy and then start putting those use cases to work. Now, what's on the other end of coming up with an insight? It's gotta get adopted, right? And so that means that you gotta put it into people's processes, into their workflow, into the tools that they are using every day because if it's off to the side and, and not top of mind and not encouraged to be used, it's not gonna get used and not create the impact that was envisioned for it. So there's a lot of people, process, technology, change management that goes on in fully adopting and benefiting from AI. And to me then it feels like everybody should be playing with this. And when I say playing, more than just fooling around, like just experimenting, trying it, and to the extent that they can invest more deeply in assembling that data strategy and building workflows around analytics and AI, that's very good because it'll set them up for a full-fledged future where, where they'll be much more data-driven.

Related to that, I wanna talk about big companies versus small companies for a moment. A lot of these big companies, they say, hey, the reason why we're doing M&A business is because we wanna get these economies of scale. We want to be able to have the resources to invest in technology and create all these great systems. And when we do that, it's going to make it extremely difficult for a small company to compete because they don't have the resources. They don't have this technology. But on the other hand, the small companies might respond and say, actually, we're doing just fine. We've got companies like Fidelity that have great technology that we can use, just like a company that's 100 times bigger than us is using Fidelity. We got access to the same technology. We've got access to the same software providers and other fintech companies out there. And as we all know, there's a lot of fintech companies out there that provide these services that are accessible regardless of the size of your company. Do you have any opinion on this idea of big companies have a big advantage over the small companies because of the technology, because of the scale, because of the cost of integrating? Or is that perhaps lessening as an advantage of the big companies just because the playing field might be leveling because of companies like Fidelity that make your tools and resources available to any size company?

I think more the latter than the former. And some aspect of this is that my sense is that there will always be big companies and always be little companies, right? And it all has to do with the competitive advantage that they're bringing to the So whereas the little company may not have all of that strength of their own technology department, they're also not burdened by that technology department, right? They may have also some niche opportunities for how they engage with clients and, and how do they attract business and all of that kind of thing. Now the big companies, of course, can come up with holistic solutions, clear the noise away from having to think about and manage technology from the advisor's perspective. And so that all makes sense. You're making a great point though, that there's a technology democratization that's playing out increasingly over time. And larger solution providers like the Fidelities of the world are definitely interested in, in helping firms, irrespective of whether they're big or small, try to meet a lot of their kind of like technology platform needs. So that firms can focus on the relationship and the advice and growing their businesses.

Let's say an advisor listening to this says, I really want to go full speed ahead on AI. I want to make sure that I'm really building a firm that is technology forward thinking here. What would be some of the typical barriers that they may have to overcome, whether it's people barriers, like you touched on this a moment ago about getting people to adopt the technology. Like, we're all stuck in our ways. We all want to keep doing what we're doing, and it takes extra effort to learn something new. So we tend to be ruled by inertia to some extent here. So that's certainly one. But what are some other common barriers that you see placed in front of advisors that want to adopt this type of technology?

As you can see, and it was part of the earlier part of our conversation, like, it may not be front of mind when you're making your tech tools and tech capabilities decisions to say, like, how does this enable analytics and AI? And so that's definitely an obstacle that the primary objective is to have a well-running business that has all the right tools at the fingertips of the associates. Okay. But once you've cleared through that, now getting the data together, getting the tools, they might not necessarily be able to speak with one another, trying to organize that. So that's not like a tech challenge that we should just make sure that we continue to recognized as part of this. You're right about, and it was the point I was making before, that there's some sort of like cultural change, adoption change, like people aren't minded necessarily towards being analytics and data and AI driven, right? And so you have to push through that adoption challenge. Even who's gonna do the analytics work becomes a question, right? And so again, that, that might be like a big firm versus small firm sort of conversation topic. Do you go to hire a business intelligence analyst, right? And do you Do you have them build up dashboards and tools and all of those data-driven insights? Or do you have somebody who's tech savvy enough within the company, maybe in the smaller firm, to be able to engage with solution providers, third parties, to identify who can help to enable me in the best way with analytics. And so that requires some analytics savvy, right? To be able to just even know how to engage it, appreciate the impact that it could create, and then bring that back back to the business so that you can go through all of the people, process, and technology adoption challenges that you're going to have to face.

Let's talk about some of the ethical considerations when it comes to AI, and let's use ChatGPT as an example. So we know that ChatGPT feeds off of datasets, data that's out there on the internet and other, any other datasets that it can consume, but they didn't really pay for that material, and now it's absorbing it. It's spitting it back out when someone asks it a question. And OpenAI is apparently making a bunch of money off of this right now. So that's an issue. Another issue is if I use ChatGPT or Bard or any of the other AI systems out there to generate content, do I need to put in a footnote that I use ChatGPT to help create this content? Do I need to disclose that I went to ChatGPT-4 to help me come up with a few questions that I'm asking you here today, Andrew? And to do some of the research. So where do we walk the line? How is all of that going to get developed over time? And where do you think that's going to land?

Yeah, this is such an important question. So as much as generative AI is compelling technology, it also comes with lots of risks. And as you're saying, like, there's such a tremendous conversation going on right now about this, and that's in the public sphere, but also we can see a lot of interest from regulators is in Congress, right? It's very early and these questions are still being worked through right now. Yes, I can recognize there are privacy concerns. There are questions about what data was used to train AI models. All of that's got to be resolved. And I imagine that regulators in Congress will not sit still about this. There will probably be some kind of action there. And like one of the things even is that the results itself of generative AI may not be trustable. We've heard about the idea that ChatGPT and other large language models, they can hallucinate, so to speak, right? And that basically means that like they give, they confidently give an answer that is just made up and is not true at all. And so it really speaks to the fact that this is nascent right now, that while it's important to experiment with it try out some use cases, work to get familiar with the technology, it's still in a place that we should have a pretty strong governance in place. We should make sure that a human is in the loop, or maybe we flip that comment and say it's a human thing and it's AI that's in the loop, right? Let's treat it in that way and do that with most use cases until we're getting a little bit more comfortable about How do we use it? How do we disclose our use of it to our clients and to our businesses? How do we monitor it for quality and, and our confidence to include it as part of our technology? That's gonna take time to sort out.

You are obviously very deep into the technology here with AI. What gets you excited? What gets you jumping out of bed in the morning? When it relates to the things that you're working on here with AI?

So Steve, how we had that discussion about how there's so much in front of us, it could feel complicated and like it's never ending. I like to do that 10-year look ahead and say like, there will be tremendous change here. And it's a great opportunity to create an impact. When I think about this and I come into work every day, I'm thinking about like, how do we enable fidelity as a business with this type of technology and how do we enable our clients with this type of technology? And we're working to identify and, and work through and put into production like real use cases that will have that impact incrementally over time. And so whether it's for ourselves about operational efficiency, finding ways to like have excellent service calls as As clients call us, we want to like sharply, efficiently, effectively answer their questions, be very proactive, right, in engaging them, offer clients analytical solutions that enable them to become proactive and to find efficiencies in their day. Right now, I feel like we're in a place where everybody needs a lot of education about what this is and what it can be. And through that education, imagine that almost as the top of the funnel, we're trying to widen that out as much as possible.

Possible.

If everybody can appreciate the technology and what it could do, they'll be able to come up with ideas for the best and most impactful use cases in their businesses. And so I and my team were listening for that. We want to hear from clients about their use cases and their pain points, because then when we prioritize those, we can put our strong data scientists against those problems and come back with meaningful solutions that are helping today, tomorrow. And so again, when you take that kind of like rearview mirror 10 years back, you see so how much transformational changes happen with technology. I think looking 10 years forward, we'll see something like that looking back on all of the AI opportunities that we were working with our clients on.

Speaking of listening to your clients, what are your advisor clients saying to you, asking of you?

What are they wanting from AI that you guys are working So definitely right now, the conversation is about ChatGPT, generative AI, what kind of use cases, tell me what this is and what it could do for our business. So we're engaged in mutual conversation with our clients that says, we understand that ChatGPT in the world is a thing that you can ask anything of and get some really remarkable answers. So let's talk through what that could be in the business setting. Setting, right? And do you need some kind of ask anything of fidelity type of capability? How would we work together to build something like that up, right? We're trying to have those types of conversations, but it always homes down into help my advisors prospect better, help me have deeper relationships, help my home office have better service efficiency, operational efficiency. It's all the usual business problems that everybody's been interested to tackle where they're saying, how can data help me with this? How can AI uniquely find patterns in the data that our associates might not be able to detect? How can we bring automation to this through data and AI? For me, that's actually really helpful. It's not some mysterious use cases that will come out of nowhere that we're talking about. We're pragmatically speaking about kind of the way that we've worked in the past and saying, how could that be different? And if it's data-informed and data-driven.

Andrew, as we wrap up here, is there anything else that you want to share or comment on that perhaps we haven't talked about yet?

Steve, it's been a great conversation so far. And so I really appreciate the thoughtful questions that you brought today. My summary on all of this is while it's a nascent capability, especially in this generative AI space, but AI in general in terms of adoption in the industry still feels very nascent. So we should all try to work to clear through that hype, recognize that there are real capabilities that exist right now and should be able to be leveraged to great impact in the way that advisors function and the way that our firms work. And that while it's still nascent and there are lots of risks and lots of scrutiny from regulators and Congress and in the public conversation, That's not a reason to completely turn a blind eye to it. It's, it's a good time to get to know this, think about your data strategy, experiment with the technology, find ways of being comfortable, compliant, and safe with your data, and use it, adopt it, because you'll find that point of differentiation.

What are some resources that Fidelity has that you could direct advisors to, whether it's PDF reports that you guys have put out or other online resources that might help people get more more knowledgeable about AI?

Absolutely. So we have thought leadership pieces that speak about AI very specifically. Steve, please check out the thought leadership on institutional.fidelity.com. That's the front page that you can access a whole bunch of products and services, including our thought leadership.

Andrew, appreciate you being on the show today. It's great conversation, and there is just so much exciting things to be looking forward to when it comes to artificial intelligence., and you've helped, uh, enlighten us along the way here today.

So appreciate it. Thanks, Steve. It's been a real pleasure today.

All right, that's all for today. Make sure you like and share this podcast through your favorite social platforms, and for more great podcasts, visit us at barons.com/podcast. Take care and be safe.

This podcast was brought to you by Wells Fargo Advisors. Looking for a firm your clients can't outgrow? Wells Fargo Advisors offers financial advisors full balance sheet solutions with wealth planning, investment products, banking, and lending that meet the needs of affluent to ultra-high net worth clients. Build your practice your way and on your schedule with independent and employee model options designed to meet your needs today and tomorrow. Learn more at joinwfadvisors.com.