E6: AI Revolution in Wealth Management: A View From Basepoint Wealth

What role will Artificial Intelligence (AI) play in the financial industry and how will it affect the role of advisors?

We tackle a great listener question related to all the AI buzz, "How might the recent developments in artificial intelligence change the financial landscape and wealth management? Will AI advisors make human advisors obsolete?

We hope you’ve gained some valuable insights or maybe even a fresh perspective on our topic today. We would love to hear from you with your questions or specific topics you would like us to cover. Simply email us your questions or suggestions to info@basepointwealth.com and who knows, your topic might be featured next.

Be sure to subscribe to be notified of upcoming episodes. Visit www.basepointwealth.com for more information and important disclosures.

Transcript

[00:00:00] Welcome to the Point Podcast. We have informed intelligent conversations about today's financial topics submitted by viewers like you. Let's go ahead and get started Here are your hosts, Landis Wiley and Allen Wallace.

Hello, and welcome to today's episode of the point sponsored by Basepoint Wealth. I'm your host, Landis Wiley sitting here as always with Allen Wallace, Chief Investment Officer. Appreciate you taking the time today to tune in. If this is your first time here, welcome. The way that this works, Allen and I love to talk about finance, and so we thought it would just be fun to invite you into one of our conversations.

The twist is we have no idea what we're going to talk about. So we've got a hat full of topics that have been submitted by you, the viewer and the folks at Basepoint Wealth. And in just a couple of minutes, we'll reach in, pick out a topic and see what we get to banter about. So, before we get into [00:01:00] that, we like to kick off each episode with a brief segment called the most important thing.

And as the title says, the most important thing is something going on in the world of finance today. Okay. They may be impacting your pocketbook, may be impacting your portfolio or may just be something that's driving economic news of the day. So, without further ado, Allen, what's the most important thing?

The most important thing right now is artificial intelligence or AI. Confidence in the future of AI is at an all time high and reminiscent of the rise of the internet in the late 1990s. Any stock remotely connected to AI is surrounded with euphoria. NVIDIA, previously the largest company in the world before a small recent correction, with a market capitalization of 3.

2 trillion dollars, is the major supplier of graphics processing units, or GPUs, and it is important to note that the total revenue of all AI applications in the United States is still only 200 billion dollars per year. The major players in AI right now are also the largest companies in the world.

NVIDIA, as the creator of hardware, and Microsoft, Apple, Amazon. com, Meta, and [00:02:00] Alphabet, as the developers of software models. Staggering amounts of money are being shoveled into capital expenditures and building of data centers to power AI technology. It is estimated that in 2024, all of these companies will spend over 20 billion dollars on AI research and development, with the largest, Microsoft, eclipsing 40 billion dollars.

Much of this will go to the purchase of GPUs from NVIDIA. Signs are emerging that faith and AI are starting to waver for two distinct reasons. First, as previously mentioned, the cost of AI is enormous. And secondly, finding ways to monetize the end product are proving elusive. While AI is fun to play with, and saves time in many applications, it still takes significant time to verify that results are accurate, and AI is prone to hallucination and simply making things up.

With interest rates rising and the cost of capital increasing, it may become much more difficult for companies to convince shareholders that these investments will eventually pay off. Any future recession could cause this entire feedback loop between hardware investments and developers to break down, which right now includes over 32 percent of the S& P 500's equity allocation.

A final note is that this entire [00:03:00] system rests on an already vulnerable power grid. Nuclear reactors are being put back online just to feed the intense hunger for power that these models require to run. There is increasing competition for power between data centers running AI, power to charge electric vehicles, and miners processing cryptocurrency algorithms.

This drain on our energy grid comes at a time when we are trying to move to more renewable and sometimes less reliable sources of energy. There is no question that AI will remain relevant, at some level, far into the future. As with most technological innovations spanning the last 200 years, like railroads, telephones, automobiles, airplanes, and the internet, it feels like the expectations greatly exceed reality.

And so much depends on this technology pitching a perfect game. Many people could see enormous losses in their portfolios if they have large exposures to AI. So keep your eye on your personal exposure to AI and be careful. Today's episode is brought to you by Basepoint Wealth. A question for you to consider.

Is your present financial strategy working for you? Do you want another viewpoint? It may be time to talk to the experts at Basepoint Wealth. [00:04:00] Call 319 826 1898 now for a complimentary financial consultation. Take control of your financial future with Basepoint Wealth.

All right, welcome back to The Point. Allen, thanks for that that segment. Always informative. Well, we're going to get here to the meat and potatoes of our, of our show and pull up our hat, figure out what we get to talk about. So always enlightening and entertaining, we hope, and a little terrifying and a little terrifying.

So, all right, let's let's see what we get to chat about today.

We need people to send in some more topics because hats run a little low here. So, all right, what do we got how might the recent developments in artificial intelligence change the financial landscape, wealth management or wealth management landscape? Well, AI advisors make human advisors obsolete.

So, obviously artificial intelligence, it's [00:05:00] been around since what was it? 1983 when the first Terminator movie came out, I think is, you know, that was my first experience with AI as you will. But also remember that's sort of the marriage of AI and robotics. Right. So AI is at a certain point.

Robotics is at a certain point. I saw the robots stacking boxes at Amazon warehouse and looks like they've got quite a ways to go before they're chasing us down the street. Right. Well, and hopefully it works out better for humanity the next time that that rolls around. So, so AI obviously been a topic of, of big conversation, particularly in 2023.

It seemed like it was all you would hear about back in January, February certainly impacted the investing world. To start the year because a number of companies had their share prices sort of shoot to the moon. It reminds me a little bit of 2000, 19, well, 1999. When, if you had the word. com in your name, your stock immediately went up by five times, right?

Anything to do with AI in this [00:06:00] market has been a little higher and it's sort of been creating this little mini bubble. You know, the, the largest companies in the S & P are. remotely tangential to AI, you know, you got NVIDIA that makes the chips and you've got Microsoft, that's sort of working on the large language models and Google and Apple and all those other companies are sort of dabbling in AI as well.

And so it's been driving the stock prices. And for a lot of people that are tuning in, maybe what they might be familiar with or more familiar with from the standpoint of AI is a ChatGPT, right? Lots of, lots of news about ChatGPT. And you know, initially when that first rolled out, I think it was kind of the, the new thing on the block and really exciting.

And so immediately everybody's thoughts went to, Oh my gosh, you know, this is going to replace my job. This is going to replace. you know, this profession or what have you. Clearly since then, there's been a number of stories that have come out about maybe some inherent weaknesses. We've tempered our expectations.

I believe there was the story of the attorney who was using [00:07:00] a ChatGPT to write his legal briefs. And when the judge went to look up all the references or the cases cited, he found that they were completely fictitious. So the thing about AI is that It doesn't it doesn't mind telling you a lie or making, making something up in order to tell a good story.

And we've talked about this before, and that's the divergence of confidence and competence. You know, the AI comes across as highly competent, but when you dig down deeper, you see that it's lacking in competence in some cases, right? And so I guess, you know, to the question of how is AI going to specifically change our industry and, you know, is it going to, you know, to a degree maybe render advisors obsolete or money managers obsolete.

Maybe that's a good place to start. Sure. Is it just has weaknesses right now. It, it, it doesn't necessarily know everything that humans at this point now let alone, you know, getting into, you know, financial advising is about more than just knowing facts. Right. Right. [00:08:00] It's about understanding people and understanding circumstances.

And so, you know, as we look at AI. where, where do we, I guess, where do we see opportunity for AI in what we're doing? So the most applicable usage for me is just in sort of organizing data, right? Making it contextual instead of having to sift through things, being able to ask questions in, in plain English and have.

Responses produced. You know, so if you have all of your data running through an AI model and you want to know, you know, how many people in this group are over age 70 and a half, and it, it can produce that immediately without having to use Boolean logic through some sort of filtering system. You can ask a direct question in plain English and receive a response.

And to me, I, so I think that sort of librarian of data is the best use case for it. The, the issue is that. All of our, all of our knowledge is. Derived either through reason [00:09:00] or experience and reason is sort of easy to get because you can download thousands of formulas from the internet. And I think that's really where AI is applicable is in loading logic and formulas and things like that.

So purely reason, but it lacks judgment. It doesn't have experience. It doesn't experience emotional responses. It doesn't work directly with clients. And so, you know, we've got hundreds of years of experience here in dealing with real problems. And so, you know, it's experience that tells you what problems you're facing.

And so that, that's the issue is trying to preload thousands of years worth of history into an LLM is, is virtually impossible. So, while you can, you can upload a lot of formulas and maybe even know how to process them, knowing when they're relevant and knowing what the exceptions are is almost impossible to program.

And keep in mind that these things are programmed by humans. So, it, it'd be. It's virtually impossible for them to know something that humans don't [00:10:00] already know. And unfortunately, a lot of things that humans know are just plain wrong. Right. Well, and you hit on something early on there that is fascinating to me as far as the use of AI, particularly in our space.

So much of what we do in working with clients is data driven, right? It's, it's getting facts. It's getting you know, data. Statement values. It's getting tax information, income information, and expense information, and then trying to take that and, and push it through essentially a series of questions to figure out what topics matter based on the facts and circumstances that are there.

One of the biggest challenges that financial advisors face is, particularly those that are handling really comprehensive planning for a client, is there's just a massive amount of data. There's a, there's a ton of data points there and knowing how to go filter that data to narrow it down to which things matter that I need to look at is a pretty arduous task.

And frankly, you know, for any one person [00:11:00] out there trying to do that, you know, as a, as a one man band, so to speak. It's fairly easy to overlook something. Whereas it seems like maybe some opportunity where AI could step in is if, if I can provide the artificial intelligence system with that same information, and then simply ask it questions in normal conversational language and allow it to translate that into the yes, no, or the binary, you know, go this way.

Yes, this applies. No, it doesn't. That, that seems like something that might be applicable. It could be, but, you know, the big issue that we have is that the context hasn't been established yet in AI language, you know, the data of a financial plan isn't nearly as important as the assumptions you can take.

data of of vastly different proportions and apply assumptions to that data to, to make the, you know, to make the financial plan look one way or the other. So, it's really [00:12:00] our rates of return. It's our inflation rates. It's our it's our expense rates. Those are the things that matter the most in a financial plan.

And unfortunately, those are art based, not science based, right? So you still need someone who has personally delivered hundreds or thousands of financial plans over time to know when certain things apply, and I just think that it's going to be many years before the AI is there yet. And maybe one of the challenges in today's AI, right, which are primarily these large language models, these LLMs, is that they're essentially just scraping data.

Right. Right? They're scraping data from the internet, which, to a large extent is forming the basis of the assumptions that it, that it is trying to make. Right. Right. And I don't know about you, but it seems to me when I go out and browse the internet on any particular topic, you know, 90 of your top 100 results all seem to kind of say the same thing.

Sure. And I don't know whether it's bias or just, you know, [00:13:00] group think. Or whatever the case may be. But you know, if, if fundamentally the assumptions that it, that today's AI systems are being built on is simply just gathering what's the most popular. opinion that's out there. That seems like it might maybe lead to some bad conclusions.

I mean, I think that's a really dangerous way to make decisions is based on popularity. You know, you, you, you're never right or wrong because people agree with you. You're right because your analysis is right. And so, yeah, I think outsourcing are. our decision making to popular opinion is probably a really bad idea.

So I think we're going to keep doing the work ourselves here until AI advances quite a bit, but in the meantime using it as a data librarian, I think is highly effective. You know, running scans over, data to see, you know, if certain triggers have been met or if certain thresholds have been breached in order to do rebalancing or in order to create cash or to know when required minimum distributions are going to start [00:14:00] without having to run separate lists in spreadsheets.

So, I mean, I think it'll definitely maybe grease the wheels a little bit, but it's stuff, it, it, you know, I don't see it as a position to replace our decision making capabilities at this point. Now, if we could take AI and do, you know, the 50 percent easiest things that we do and then spend all of our time on the 50 percent hardest things that we can do, I think that would be a great use case for it.

But you got to keep in mind that you can go back to the to the, the Luddites of, you know, 1811 to 1816 trying to destroy all the textile mills because, you know, their jobs were going to get replaced. And that trend had started in the late 1600s with, you know, with the loomers and the, and and all the other textile type manufacturers.

So we've had this natural fear of replacement for hundreds of years. And what we find is that when we hold back progress, we're doing just that. Letting Automation or machines or, you know, equipment do [00:15:00] mundane tasks is allowing the rest of us to move on to higher level activities. And so I don't want to stand in the way of progress because of fear.

Now, there may be legitimate fears, like what do you give the AI control of? You don't want it to be able to launch missiles. I mean, you know, I think that's probably just common sense, right? You and I had talked once about the report of the, the training missions they were running with the idea.

Remember this? And it had some sort of restriction of, it couldn't do you, do you remember this story? Yes. And so it came up with a solution to kill the operator because it was holding it back from its mission, right? Which is the classic Hollywood narrative around AI and robots, right? Is it fundamentally, they figure out that, you know, humans are the biggest

threat to humans. And so therefore you have to kill the human to protect. Yeah. So, I mean, they found this in real life in a real life example. Now, luckily it wasn't actually connected to any real military equipment, but at least in the simulator, you know, it started trying to, you know, [00:16:00] wipe out the person operating it because it was holding it back.

And then it started trying to destroy the communications towers so that it could pretend like it wasn't getting the message. So, yeah, I mean, I think, I think in the wrong hands, we could definitely be dealing with something dangerous here, but, you know, so, so I think regulation really is necessary, but it's really more about controlling what you hook it up to as a, as opposed to holding back progress and trying to save, you know, menial jobs in the, in the present.

Well, I think that's a great point on AI. To me, it just seems like the latest sort of innovation in technology, right? And to your example of the looms and the textile mills, that was technology, right? Maybe not in the way that we talk about it today, right? We associate technology to computing systems and what have you, but really technology, you know, the wheel is technology, right?

The banging a rock on, you know, something is, is technology. I can remember coming up you know, in school as the internet was, you know, just sort of becoming a thing. And, and, you know, the fear [00:17:00] at the time about everybody's jobs were going to be replaced because the internet, okay, have jobs changed in the last 25, 30, 40 years, as the internet has, has grown and become more pervasive.

Absolutely. They have, but to your point, a lot of the things that have been replaced are the, the menial. Right. Right. That were time consuming. It's probably not a coincidence that some of the fastest, most explosive advancements in technology that have inarguably made people's lives better and more effective have happened really in this computing era, where computers have incrementally taken over more and more of the mundane.

Right. tasks that sucked up our time. Well sure. Yeah. I mean when I was in my late twenties, I went to a dentist office where they were still using a typewriter to, you know, type out little cards, and put them in a paper file. and I thought to myself, man, I haven't seen this happen anywhere else. Right. Like it was so outdated, but then, you know, it may, it [00:18:00] got me thinking about how much time they were wasting by, by physically shuffling pieces of paper around and using a typewriter and how much storage space they were taking up, you know, with these little cards that they were managing.

And so, but I do think that there's been a great reduction in the amount of paper that we use, the amount of loss that we use, the amount of illegibility that we, that, You know, that we face and all those different things. So, you know, I, I do think that computerization has greatly enhanced the efficiency of operations.

Now, I think in some cases it's degraded customer service. I think it's, it's allowed more things to get lost, you know, it's there's something imposing about a file cabinet full of physical paper, but you can put that stuff on your computer and never see it again, right? So, but I think there's always a cost to progress.

And so we, we want to make sure that we maximize the benefits of these types of things. I think AI will definitely benefit us over time. I don't think it's something you should jump onto as an investing bandwagon. I think it's way ahead of its time right now. I think it's still in [00:19:00] sort of the the carnival phase.

You know, it's, it's more of an interesting trinket to go on and, you know, see if it will do your homework for you, that kind of stuff. We have to be really careful about the results that you get because the results are spurious right now at best. Right. And from a impact, you know, to the financial advising profession, what it maybe seems it'll present the opportunity to do over time is for advisors, you know, servicing teams, investment teams, you know, all the folks that are involved in helping people organize their financial lives.

I for one would be incredibly happy if I could spend less time moving numbers around and more time really digging deep into the problem solving aspect for a client. Absolutely. And I think the majority of our clients right now still appreciate that human interaction. It, the robot doesn't care about your emotions.

It's, it's not in the business of trying to mitigate your fears. And so unless you program that in, you're not going to get that type of treatment. So, you know, around here, we [00:20:00] understand your feelings. We understand your goals. We understand your thoughts. They're not just abstractions to us. We've helped hundreds of people, thousands of people reach their goals over time.

And we're going to continue doing that until So, you know, I, I'm not afraid of AI, I, I'm happy to implement it in areas where it's useful. Making sure that security is, you know, is, is up to speed. But I, I think we can definitely get a little bit of leverage out of it, but I don't think it'll be sitting at my desk anytime soon.

So maybe the, the message here, you know, to close it out, you know, to answer the question is certainly there's opportunity. It's probably very, very, very early just in, in technology in general. Is it going to change the landscape of wealth management? Probably. Hopefully in ways that are going to benefit the client by allowing us to spend more time, as you said, doing the things that are really chiefly in the domain of a human.

Right. Right. Leave the rest of it to a computer. Right. Yeah. I mean, I've heard it said that we always way overestimate what we can get in a year and we always [00:21:00] underestimate what we can do in 10 years. So I think that for the next few years, we just let this play out, but I think when we wake up 10 years from now, we're going to be surprised at how far this stuff has evolved.

And I think the same thing happened with the internet, you know, 2001 we were, we were anything with com in it was, was blowing up Toys R Us, you know, or Toys. com was, was valued at like 10 times Toys R Us, even though they didn't have any physical location or any actual assets and they had a, a business model that was losing money.

You know, we want to be careful to avoid those type of traps. Right. Perfect. Well, I think that's a good place to kind of wrap this one up. So good discussion, certainly something that we'll pay attention to. and take advantage of as technology evolves. So thanks again for tuning in, taking some time out of your busy day to join us here on The Point.

Again, if you've got a question or a thought that you'd like to hear us chat about up here, feel free to visit our website, Basepointwealth. com or send an email info at Basepointwealth. com with your question or your [00:22:00] discussion topic. And who knows, maybe we'll chat about it. One of these days. So until next time, take care.

Thank you for joining us for this episode of The Point podcast sponsored by Basepoint Wealth . As always, you can submit questions or topics you'd like to hear discussed to info at Basepointwealth. com. Be sure to subscribe so you don't miss any future episodes. Basepoint Wealth LLC is a registered investment advisor.

Information presented is for educational purposes only and does not intend to make an offer or solicitation for the sale or purchase of any specific securities, investments, or investment strategies. Investments involve risk and unless otherwise stated are not guaranteed. Be sure to consult with a qualified financial advisor and or tax professional before implementing any strategy discussed herein.

Past performance is not indicative of future performance.