The Dividend Cafe - Artificial Intelligence and The Bahnsen Group

Episode Date: June 21, 2024

Today's Post - https://bahnsen.co/3KSRGPp Navigating AI Investments: Lessons from the Dividend Cafe In this episode of the Dividend Cafe, recorded live from Atlanta, Georgia, David discusses the topic of artificial intelligence (AI) and its impact on investments. He explains why AI hasn't been a major focus of Dividend Cafe, despite its prominence in media and financial conversations. The discussion delves into the distinctions between AI's backbone companies and those utilizing AI, drawing historical parallels to the internet boom of the 1990s. The episode outlines ten principles to understand and navigate AI investments, advocating for a dividend growth framework. The host addresses market concentration risks, historical lessons, and the sustainable approach for investing in AI-related companies. Listeners are encouraged to view informative charts at DividendCafe.com and reflect on a quote from Charles Mackey about collective and individual thinking. 00:00 Welcome to Dividend Cafe 00:02 Travel and Speaking Engagements 00:43 AI in Financial Media 03:44 Investment Implications of AI 05:18 10 Key Takeaways on AI 07:39 Historical Context of AI Investments 13:21 Predictions and Pitfalls of AI 17:04 Principles of AI Investing 19:16 AI's Broader Impact on Society 20:53 AI and Dividend Growth Investing 24:59 Conclusion and Final Thoughts Links mentioned in this episode: DividendCafe.com TheBahnsenGroup.com

Welcome to the Dividend Cafe, weekly market commentary focused on dividends in your portfolio and dividends in your understanding of economic life. Hello and welcome to the Dividend Cafe. I am recording live in Atlanta, Georgia, where I had a speaking engagement this morning, got to spend a little time with some clients, and then am speaking at a large conference on Saturday before heading up to Grand Rapids, Michigan, where I will also be with clients next week and speaking at a couple of events, participating in a large symposium that I speak at every year. So I'm in the midst of a little bit of travel away from New York and away

from California. But nevertheless, here in Atlanta, I have created a dividend cafe that was not actually on my agenda this week. Initially, the subject of artificial intelligence is not exactly being ignored in the media, you may have noticed. And sometimes that does take on the flavor of addressing how much Nvidia stock is going up, for example. And other times it has to do with all the conversation and reporting around a sort of ecosystem of stocks that are involved in AI to some degree. Other times, there's a lot of discussion around what AI will mean for the economy, how much of our current productivity is dependent on AI, how much the current market is dependent on AI. And in circles that I traffic in, there's a fair amount of conversation, even aside from the investment implication, on what the cultural or ethical or political implications

of artificial intelligence may prove to be. But I bring all that up just to simply say that this is not a topic I've ignored in Dividend Cafe. It's just not a topic that I write about week in and week out for the very simple reason that everybody is talking about it. And I think that what you get a lot in financial media is similar to what we get a lot in other aspects of culture, which is stuff just getting overdone, oversaturated. Particularly with financial media, if there's a chance to scare people a lot, the newsletter industry is almost entirely driven around trying to play into or exploit people's fear. But a lot of financial media too can, more conventional media, newspaper, television, broadly watched, broadly

viewed internet sites center around people's, I don't want to say just greed, but the hype, the desire to believe that a really big investment return can come easily. The frustration that comes from believing that your friend is or neighbor is making money that you're not making. These things are at the core of the business model of financial media. And Dividend Cafe really doesn't exist to go into those elements of excessive fear or excessive hype that already circulate and dominate other elements of message delivery. Our content message has a particular vision and it's not always that exciting. Grateful for the audience it does have, and I hope that it is scratching the itches of those that either read Dividend Cafe or those of you that watch

the video or listen to the podcast. Obviously, I care about the message being useful, but it isn't meant to be trendy. And AI is a trendy topic right now. And so it just isn't one that becomes the epicenter of our efforts here at the dividend cafe. But I would say that in an investment context right now, I have implicitly been talking about AI quite a bit lately because I've been talking about the top heaviness of the market, for example. I've been talking about the high concentration for index investors in exposure to just a few different names that are either directly connected to a sort of AI narrative or are really heavily adjacent to an AI narrative. And so even though I think that's a more practical investment lesson, it may not seem to be a total AI conversation or narrative. And yet I think it is really one in the same that we're discussing the relevance of the current artificial intelligence moment on a broader level of investors. Today, I want to dig a little deeper than that. Rather

than just merely talk about what I think are the embedded risks that have materialized for index investors because of this high concentration around AI, I want to just lay out a number of different principles that represent the Bonson Group's belief about the artificial intelligence moment. Some will say, how come you guys aren't invested in AI. And we would say we absolutely are invested in AI. And we can show you very particular companies and strategies as to where that exposure comes from, how it has been monetized already, what that looks like going forward, the way we're thinking about all of it. But even apart from just a particular, here's on my statement where you can see that we own AI type of thing, I created for our purposes in today's Dividend Cafe, 10 kind of takeaways or summaries

of our thinking about the total artificial intelligence moment. I'm going to just go through these in order. They're listed out at dividendcafe.com and we'll elaborate on each as we go. And hopefully it'll be something useful in terms of getting a better understanding of how we're thinking about the AI moment. Number one is that tons of what we're talking about with AI is not new. It's been around for some time. Now, I do recognize that a lot of generative AI is newer than some of the particulars in predictive AI. And some things that I would consider a byproduct of the vast amount of research and also investment into AI wasn't really called AI then. But various components of language learning models are not totally brand new. The predictive elements around analytics have been around for some time, and they've gotten faster, they've gotten more sophisticated, and they will continue to do this entire kind of concept of what they call multimodality, I think is reasonably new in the landscape

of how this conversation has come together. is reasonably new in the landscape of how this conversation has come together. So the promise of the technology of AI has definitely grown immensely. But I just think it's important to remember that you did not get hundreds of billions of dollars of investment. Like the GDP of some countries had gone into AI in the last 24 months or 18 months or 12 months. This is in some cases decades in the makings. And a lot of the application of it now is newer. And that's the moment in which we find ourselves, particularly when we think about the promise of generative AI. Number two, and now I think some of these things are

going to become really historical and useful, and I'm excited to unpack it. Number two is that nearly all of the investment benefits of AI thus far have been limited to what I'm going to call the backbone of AI, not the use of it. By backbone, I'm referring to the infrastructure, the tooling, the things that help make the use of it possible. So companies that are using or doing AI is very different from where the investment gains have been so far, which are the companies that help to make AI which are the companies that help to make AI possible. This is largely chip companies so far. And then I would argue that a pretty fair secondary example would be in certain cloud companies as well, most notably Microsoft.

But nevertheless, the investment gains have largely been about the infrastructure of AI so far. Number three, many of the backbone winners will prove to be losers. And this is where the history now starts to become very fascinating. There was a very similar dynamic with the advent of the internet in the 1990s. There were certainly dot-com companies going public and becoming famous and running ads at the Super Bowl and doing celebrity endorsements and becoming pop culture phenomenas. But they also really were engaged in something that I think was very, let me put it this way. The dot-com moment was largely

backbone driven until it wasn't. The telephone companies, cable companies, wireless network companies, and then a lot of adjacent products and services that went around to that from servers, routers, et cetera. You look at the history of the 1990s, there are companies, when I say backbone winners that proved to be losers, companies that are gone altogether, like Lucent and Global Crossing, companies that are still around, but essentially are just nobody companies, Juniper Networks, companies that went away altogether. Your son, Microsystems and Junipers and Worldcoms in the 1990s were deemed to be the new big things necessary to go power the internet. And it wasn't that the thesis of backbone companies was wrong.

It's just some of those backbone companies ended up proving to be losers. And I don't think this is a byproduct of anyone who didn't appreciate that at the time was anti-technology or not keeping up with the times. Warren Buffett was famously very demonized for maintaining a focus on kind of value investing. But there also is just a sense in which there were certain things that were unknowable. And that is very much the case now, in my opinion, around the AI moment relative to the way we saw this unfold with the internet. Now, number four is that many of the backbone winners that will stay winners are already priced for such. This is where a company like Cisco from the 1990s and NVIDIA now becomes an opportunity for me to make an analogy. I've done this already multiple times. We've put the chart out in some more recent writings of ours. It's available again

today at dividendcafe.com. But it's not just simply the Cisco analogy. You had companies that are today huge winners of what took place with the internet 30 years ago and 25 years ago, etc. And yet, in almost any case you could think of, experienced massive shareholder losses, experienced massive shareholder losses, meaning stock price drawdown that in some case never fully recovered, in some case took well over 10 years. And so it's just not as simple as identifying what the backbone sector will be or who the winners will be, who the survivors will be, because now the pricing itself has already reflected that in some cases, certain companies are going to be the significant player in whatever happens with AI, and that's been priced in and then some. I think the charts that we use to make

this point in Dividend Cafe this week are worth looking at. And I think that the economic and market lesson of what took place there is not something I ever shy away from admitting that I am influenced by. It is impossible to not have found that message for a professional money manager out of the 1990s into the last 25 years I've been professionally managing money to have not found it to be a potent message that a company can succeed as massively as Microsoft and Cisco did and yet suffer the way they did in terms of stock price as a direct byproduct of what was simply happening with the pricing of their stock before. I just mentioned two companies as a random example. I

could mention a dozen, and that's just in the realm of ones who actually made it, who flourished as active operators in the business. There's a whole graveyard of other companies that also participated in a big run-up and didn't make it, that are now basically either gone or irrelevant. Number five, predictions about how AI will change things, who will benefit from the change, and how this will all play out will be riddled with error. This is not merely a byproduct about it's going to be hard to pick who the backbone winners will be. That was what number four and three were about. But now you're getting into kind of other components that require certain presuppositions or assumptions in how we think about this story playing out that are going to open

ourselves up to fallibility. An example I use from the internet era was the notion that less and less people now you can get wireless through a mobile device. So you saw a huge downward pricing and PCs and a huge escalation in the pricing of mobile phone units. And we remember the names of Motorola, Nokia. And it's so funny to think about because Nokia and Motorola were not killed by the iPhone. Then BlackBerry comes around and then iPhone kills BlackBerry. And there was a lot of creative destruction going on sequentially. Certain companies, you talk about Darwinian business model, a business cycle. Some companies were killing others and then getting killed. And this was happening rapidly.

And you think about it now, was that laptop theory wrong? Gateway is gone. It's a multi-billion dollar company that in 2007 got sold for $700 million and then is totally gone now. So it goes down 90% in value and then gets swiped up, wiped away. But then Hewlett Packard right now is a $40 billion company. It's not at an all-time high, but it's very near it. So the laptop theory itself wasn't fully understood or executed well. And then you look within the phone component, people were

right. The advent of wireless technology was going to totally change the rules of the game around phones, but that didn't stop Ericsson and Nokia and Motorola from joining the graveyard. And that's the point I'm making is that there's going to be a lot of opportunities out of AI to get certain sub-narratives wrong and to get specifics within sub-narratives wrong. And that's been the lesson of history. It's going to be very humbling in my opinion. lesson of history. It's going to be very humbling in my opinion. Number six, greater fool theory applied to artificial intelligence investing will make fools of those who employ it. What is greater fool theory? It's the idea that the underlying investment merit doesn't matter, that as long as you have someone who's a greater fool than you are to sell it to, who cares how good of a company it is or who good, who cares how good of an investment piece of something is, as long as someone else believes

it's good, that idea of being able to sell, to move your hot potato to someone else at a profit, that's really what the game is about. And obviously greater fool theory works until it doesn't, as long as you buy something at X and sell it to someone else for much more than X. Again, the key word being that you sold it to somebody at much more than X. But greater fool theory is by definition not sustainable. And eventually the music stops playing. I think in this particular case, those who are holding on for dear life and believing that they're going to find the last sucker to sell something to, as opposed to being the last sucker, I think they have another thing coming. Greater fool theory is a foolish theory of investing.

Number seven, principles are what you apply in terms, in times of decision making, not the things you end up with out of decisions. You don't make decisions and then say, okay, so these things now become my principles. You have principles that drive the way you make decisions. FAT oriented, popularity oriented, momentum oriented decision making is not principle driven. And I suppose someone could say my principle is that I'm just going to change whatever I do based on the weather of the moment and trying to adjust around it. Perhaps that could be one's principle. History has not been kind to those who have practiced such. Number eight, speaking of history, this time is different is what people always say before it ends up not being different. And this is always a very

difficult lesson to remind people of in investing because you only get to point out that, oh, look, this time wasn't different and this thing ended up happening and so forth and so on. You only get to point out that, oh, look, this time wasn't different and this thing ended up happening and so forth and so on. You only get to do that with hindsight. Until such and such a thing plays out, it does feel like this time may be different. But the problem with this time is different is that it isn't ever different and that you are delaying the inevitable. Obviously, some companies succeed. There's nothing about this time is different that says some companies can't go from very small to very big. Companies grow into their valuation. They exceed expectations. They evolve. They innovate. There are success stories all over the place, including

some that we on occasion would miss. However, the part I'm referring to is the belief that substantially overstretched evaluations that draw in the hysteria of the population, that those things don't end up with a reckoning, that this time is different, that I am unaware of an exception to the rule in history. Number nine, the only thing that will prove more wrong than predictions about AI investing are predictions about AI's impact on society. I am a Shumterian to my core, the understanding about creative destruction as a vital, inevitable, and ultimately fruitful part of a market economy is very much at the core of my belief system. And yet I think that when we look at conversations and sometimes hand-wringing that is going on about

AI right now, it confuses the macro for the micro. There will be some individuals who lose their jobs because of AI, but on a macro and broader based system, it will create more jobs than it loses. And I'm thoroughly convinced of this. I'm thoroughly convinced that various fears of sort of dystopian consequences are part of the same category people have been afraid of with technology forever. I think there will be bad mistakes made. I think there will be egregious violations of ethics that take place along the way. There'll be law breaking. There'll be some regulation. There'll be some re-regulation. Some of it will be good. Some of it will be bad. This whole thing will play out in a certain way, but the predictions about AI's impact on society right now and what its ultimate utility will prove to be, I think, are going to for clients of the Bonson Group.

Artificial intelligence can be invested in within a dividend growth framework. In fact, it ought to be invested in within a dividend growth framework. My view is that trying to be immunized as much as possible from the lessons I talked about earlier regarding the fallibility of how a lot of these things will play out, the fact that there are backbone companies that are different than utilizers of it, that there are adjacent companies that will get caught up in the moment and some will fail and some will succeed and some will succeed, but already be overpriced. There's all these sort of historical and economical realities at play. And to be as reasonably immunized from that as possible, to attach an artificial intelligence

investment framework to cash flow strikes me as a very prudent idea. Cashflow strikes me as a very prudent idea. Having, as we do in our own portfolio, multiple companies that stand to benefit in either a backbone or adjacent backbone capacity to AI, yet without the risk of an 80% drawdown because it is tangential and not the core of the entire business, I think is a very sensible way to be engaged.

And then ultimately, to actually invest in companies that will use AI as a tool, but don't make AI as a backbone provider. The analogy that I used in the Dividend Cafe written this week that I don't think I've shared here so far in our recording is if AI were food, you're talking about right now, all the investment being in the people who make the ovens, not the restaurants. Ultimately, AI has to actually be a restaurant. If it's food, it has to get served out in some way. It has to be used and be a meaningful component. And there is, and is going to be utility around the ability of generative AI to drive efficiency and productivity in an actual real life business. But right now, when people are talking about, oh, I'm invested in AI,

and they're referring to a chip maker, for example, they're talking about the oven. And that is not the heart of the restaurant. I still think the AI moment is going to be in a lot of the basic companies we own, that if I were to say the name of them, you'd say, what does that do to AI? And it would say the way in which they're utilizing AI in driving a different impact and result and process in their business. It's a point I've made before, but you have to understand that all of us today who use the internet on a point I've made before, but you have to understand that all of us today who use the internet on a daily basis as I do use social media, you don't own AOL and Yahoo and

Lucent and CompuServe and Juniper in your portfolio anymore. The ways in which the internet has now become vital in the way we do business is evidenced in every company you can think of from a railroad to a utility, to a restaurant, to a financial company, et cetera. Across all the sectors of the S&P, the internet is being utilized. That's the way I want to think about AI. So it is untrue to say that a dividend growth methodology will be uninvested at AI. It will be invested in the application of AI, the delivery of food, not just merely the manufacturing of ovens. And yet along the way, there is still a backbone AI investment with companies that do not end up representing 80% of the portfolio, a smaller amount, but are far more mitigated against the risk of excess and valuation that I've talked about earlier. So I hope that makes sense. I want to be invested in companies who use AI to the betterment of their own profits. And those profits will come

to our clients in what we call dividends. That is the way we view AI investing as compatible with dividend growth investing. I appreciate you bearing with me on all 10 of these lessons. I mentioned there are some good charts at DividendCafe.com as always, so I'll refer you to that property. And I'll close with the quote of the week in this week's Dividend Cafe from Charles Mackey. Men, it has been well said, think in herds. It will be seen that they go mad in herds while they recover their senses slowly and one by one. Thank you for listening. Thank you for watching. And thank you for reading the Dividend Cafe. Have a wonderful weekend. The Bonson Group is a group of investment professionals registered with Hightower Thank you. information referenced herein are from sources believed to be reliable. Any opinions, news, research, analyses, prices, or other information contained in this research is provided as general market commentary and does not constitute investment advice. The Bonser Group and Hightower

shall not in any way be liable for claims and make no express or implied representations or warranties as to the accuracy or completeness of the data and other information, or for statements or errors contained in or omissions from the obtained data and information referenced herein. The data and information are provided as of the date referenced. Such data and information are subject to change without notice. This document was created for informational purposes only. The opinions expressed are solely those of the Bonson Group and do not represent those of Hightower Advisors LLC or any of its affiliates. Hightower Advisors do not provide tax or legal advice. This material was not intended or written Thank you.