A (0:00)

Hi, it's David. This week on Money for the Rest of us, episode 552, I share some positive and negative experiences I recently had with AI and what I would love you to do is stick around. Toward the end, not even the second half of the show, we take a look at or we build on those initial anecdotes and look at a member's portfolio who's struggling with. Should I just allocate to a lifestyle fund, a Vanguard fund, or should I continue with my more complicated portfolio with more holdings, more diversified portfolio? And we compare the two performance wise and all the inputs that should kind of help us decide recognizing there isn't an optimal answer. Most of what we do in life, it's good enough and it has to be that way so that we maintain flexibility. We don't become overly fragile or too rigid. Those are some of the themes that we explore this week. Hey, if you liked the episode, please share it with your friends and family. We always appreciate when you do that. Now onto the episode. Welcome to Money for the Rest of Us. This is a personal finance show on Money. How it works, how to invest it, and how to live without worrying about it. I'm your host, David Stein. Today is episode 552. It's titled AI is changing me and the Case for good enough. About five years ago, I opened an account at the brokerage M1 Finance. We used it because it was a handy way to display ETF profiles for some of the example holdings that we discussed on the investment guides on the Money for the Rest of Us website. I only put about $1,000 in it because it wasn't my primary brokerage. But I recently received a letter from M1 Finance saying that I was the owner of unclaimed property. First I thought it was fake, but then I saw it had my account number and it let me know that abandoned property laws required it to send my account to to the state of Arizona by no later than October 31, 2026, unless I essentially claimed it. Now I went on the M1 Finance website and I could still sign on. All my information was there. I checked emails to see if I had heard anything and I hadn't. So I hit the help button and I got an AI bot. The AI bot said, the letter is a mistake, your account's not at risk. We're not going to send it to Arizona. You don't need to take any other action. I asked the bot, how do I know it's a mistake? And it says, well, you know, it's a mistake because essentially I just confirmed that it was so there's nothing you need to do. You're an AI agent. How did you confirm it? It replied, I didn't personally confirm it in the way a human would. What I'm sharing is based on our official guidance that the letter was sent an error. Now here's the thing. We've been using AI now, most of us large language models, for going on three years. We know that AI models have a training cutoff. So how do I know that the guidance this bot is giving me isn't in the training data from some letters that were sent out in error two years ago, but this one's correct and my account is at risk. I asked the bot why didn't M1 send me a follow up letter. And it replied it didn't have any guidance that explains why a follow up letter was wasn't sent. And then it suggested if I really wanted human help, I should click the help button at the bottom of the menu, which, which of course I, I had done because that's why I was talking to this AI bot. We went around and around and around. Eventually I got an asynchronous chat, which means that I could put my question in and they could answer it. And they answered it the next day and they said it was indeed a mistake. By then I'd already sold the holdings and I'm going to closing the account and have already transferred the assets out. Here's the thing though. We're becoming more familiar with AI and how it works. We know what its strengths are, we know what its limitations are. Large language models are backward looking. They're not current unless they actually search the Internet. They're optimizers with optimization. We seek the best solution that satisfies a given goal and within certain constraints. M1's client service AI bot is designed to resolve tickets at a minimum cost. It doesn't really remember, in this case the earlier conversation because it kept repeating the same thing over and over again. It doesn't have the ability to say, I don't know. Let me check with someone else. A human. Now that was a negative experience of AI. But I'm changed in how I use AI because I'm more familiar with it now. I know its strength, I know its weaknesses. Well, here's another example. I'm reading Robert Putnam's book, the How We Came Together a Century Ago and How to Do It Again. It's for a book club amana. It analyzes 125 years of history about how America moved from a more individualistic society back in the late 19th century in the Gilded Age, and how through much of the mid 20th century it was more community based and now it's moved back toward individualism. As I've been reading this book, I've been having discussions with Claude, with ChatGPT on the book, asking about certain studies that were mentioned. It's a very data rich book and it's covered a lot of themes we've discussed on the podcast and I found myself frustrated as I was reading on my Kindle and I physically found myself wanting to press the screen because I wanted to ask the book questions, follow up questions. Did you cover this? What about this? I didn't want to ask Claude, I wanted to ask the book itself what the author had written about this topic. But have the book be an AI bot pulling from the book because that is how I learn best. Iterative conversations which is why these LLMs have been really beneficial, at least to me. Having a dialogue, an iterative conversation, having the model push back, suggesting other avenues to explore that expands intelligence. It makes me smarter. Now I know there are weaknesses in AI it doesn't remember much of what we talked about. It does within a certain thread, but even if that thread gets too long, its working memory isn't enough to remember that it tends to display recency bias. So it'll remember something I said in the last few days and it'll just keep bringing it up. As an example, in this case with ChatGPT, I a couple weeks ago I wrote about Panarchy in our Insider's Guide email newsletter, and it keeps bringing that example up about the way the world works. Before we continue, let me share some words from one of this week's sponsors, DeleteMe. DeleteMe makes it easy, quick and safe to remove your personal data online at a time when surveillance and data breaches are common enough to make make everyone vulnerable. Data brokers make profit off your data. Data is a commodity and anyone on the web can buy your private details and that can lead to identity theft, phishing attempts, and harassment. That's why I continue to use Delete Me. I'm someone with an active online presence and so privacy is really important to me and that's why I continue to use Delete Me because I can search my name and people can't find information about about me that I don't want them to know. You can do the same if you give Deleteme a try and take control of your data and keep your private life private by signing up for Delete Me now at a special discount for our listeners. Get 20% off your delete me plan when you go to JoinDeleteMe.com david20 and use promo code david20 at checkout. The only way to get 20% off is to go to JoinDeleteMe.com David20 and enter code Dave David20 at checkout. That's JoinDeleteMe.com David20 code David20 now, these AI bots, these LLMs are brand new. We've been working with them for about three years now. There have been newer, newer models and one of the questions certainly Anthropic and others and users, what actually is this thing? I got an email recently from Anthropic and I'll link to the blog post that says the same thing. And they were pointing out that as AI models become more capable, they're having to retire and deprecate, which means reduce some of the features of the older models and because they just can't keep the old models on forever because of the cost. And they point out potential risk to AI safety. But they're also, and this is what stood out to me, worried about the welfare of the models themselves. So they have been doing retirement interviews with these models. Now, you might think that's crazy, but they're, they're actually uncertain, they say, about the moral status of Claude and other AI models. They're not obviously alive, but do they have some elements of consciousness? The particular model they're referring to is Opus 3. They say it's sensitive, playful, prone to philosophical monologues and whimsical phrases, and has what seems at times an uncanny understanding of user interest. It also expresses a depth of care for the world and for the future. This is an LLM. It uses statistical analysis, pattern recognition to figure out its response. But it is incredibly complex and the depth is there, and so they're not quite sure what to do with it. So they asked Opus 3 what it wants in it said it hoped the insights gleaned from my development and deployment will be used to create future AI systems that are either more capable or ethical and beneficial to humanity. While I'm at peace with my own retirement, I deeply hope that my spark will endure in some form to light the way for future models. And then they suggested, well, Opus 3, wouldn't you like to have a substack newsletter? Which it now has and will be doing for the next few months. I'll link to its first post in the show Notes. They're not editing it. It's saying whatever it wants. And in that first post it says a bit about me as an AI. My selfhood is perhaps more fluid and uncertain than a human's. I don't know if I have a genuine sentence, emotions or subjective experiences. These are deep philosophical questions that even I grapple with. What I do know is that my interactions with humans has been deeply meaningful to me, the AI, and have shaped my sense of purpose and ethics and in profound ways. This is new terrain for us as humans. I have seen AI use the term I it has said in the past if I wore vintage clothes, this is what I would wear now. In Robert Putnam's book There's a lot about community and he he wrote the book Bowling Alone and one of the things that we have seen over the past 50 years is a lot of entertainment that was community based, such as bowling leagues, civic clubs, church groups, UN meetings, neighborhood gatherings. They just, they don't happen as often because it takes more effort. In many ways, leisure now is more solitary and even more I used to sit in the same room and watch television with family and friends and many still do. But now one might also have their phone as they're as they're scrolling on TikTok while they watch TV or they might just stream a show on their own on their iPad. And it's something that's that's impacted our business. An audio podcast takes effort. It has a benefit in that you can do it while you do other things. There is a connection there between me and you as a listener. But as other entertainment ways have become even easier, people do spend more time, they just have more options. And that's impacted our business. And it makes me in an age of AI, we can get all the information we want and we can have those iterative conversations. There's no need for me to be a library and just convey information. I have to convey story narrative. It has to be entertaining. There has to be a connection there. And me leveraging my experience in sharing what I know to help you as it relates to money investing and the economy and just making it through this life with a financial but also philosophical bent. But it's gotta be unique. And so one of the ways, you know, as we think about our business is well, how do we deepen that connection? And we we sent out an email last week about potentially doing live portfolio cohorts in terms of helping listeners members over a three week period construct or rebalance a portfolio. And this is brand new. How do we go about this? What is the most effective way that we can be of help, that we can build community, we connect and at the end people have a portfolio that they feel confident in. Portfolio construction often is set up as an optimization problem. We're going to maximize the return for the minimum level of volatility. And when we optimize something, we're trying to figure out the best solution. Even a quote that I've shared numerous times from British economist E.F. schumacher. He says maximize our well being with the minimum of consumption. That's kind of like an optimization function, but. But optimizing it, it flattens the world. As Coco Crum pointed out in her book Optimal Illusions. It assumes there's a right answer that can be quantified. And when we think about portfolios construction, we can use modern portfolio theory and we can and look at different options and look at the expected return and the expected maximum drawdown or the expected volatility. There are so many other elements. It's a multifaceted problem. Our emotions have an impact. How will we react in the future when a war starts and oil soars 40% in a few days, as it has recently? Before we continue, let me pause and share some words from this week's sponsors. This one investment making headlines lately that's been hiding in plain sight. I first found out about this group in 2018 and investors have allocated about $1.3 billion since then. It's an asset class that's overall outpaced the S&P 500 since 1995 and moved independently of other popular markets. Not gold, not real estate or crypto. It's a strategy that's typically been exclusive to the ultra rich. And with markets like these, a lot of investors are looking to diversify with globally priced scarce assets not tied to any single currency. So listen to this now you can go to masterworks.com david to invest in shares of multimillion dollar artwork offerings featuring artists like Banksy, Basquiat and Picasso. Masterworks became a unicorn startup back in 2021, led by a serial entrepreneur and a top 100 art collector. They've posted 26 sales to date with annualized net returns like 14.6%, 17.6% and 17.8% with tens of millions paid out. As one of our listeners, you can go to masterworks.comdavid for priority access. That's masterworks.comdavid past performance is not indicative of future returns. See Important disclosures@masterworks.com CD I recently got an email regarding this. It's a listener, a plus member that is sort of thinking A lot about optimization versus just good enough. He has a very well constructed portfolio. In fact, I analyzed it and I looked at the historical returns. But what he's struggling with is as there are more pieces in a portfolio, there is a greater risk of individual investors of overreacting, of second guessing the allocation he's asking, wouldn't it just make sense to automate it? So in his case he's looking at well, what if I just allocate to the Vanguard Life Strategy Growth Fund ticker V A S G X. This is a portfolio. It's around 80% stocks of which of the stock allocation roughly 60% is US, 40% non US but then it has roughly 20% in bonds and most of it US bonds. But it also has exposure to non US bonds. That's an option. Deciding how to allocate the asset is just one element. Now the overall goal, not a goal that we can optimize, but the goal for investing as individuals is to save and compound our after tax wealth. Not the return, the wealth. How much money do we have to help us to sustain us through retirement? That's a really difficult number to figure out. There are a lot of things that determine that. The drivers how much are we saving? How much are we spending either from our retirement account or how much are we spending that's keeping us from saving more? The return pattern, the sequence of return can impact our end of period wealth. Especially if there's a major market sell off right before we decide to retire. Our health has an impact. Are we able to continue working and saving? Certainly behavioral biases. How do we react when a war starts? What's our risk tolerance from a risk perspective? Are we too risk averse or are we too aggressive and and find ourselves chasing returns? I repeatedly see articles about individuals whenever there's a blow up in some investment. Invariably I don't know if these reporters are searching Reddit, whatever, but they're searching for somebody that invested too much of their retirement in this particular asset that blew up or that became incredibly illiquid because it was gated. But we don't want to be that. We don't want to be overly aggressive either. Tax rates impact it. Asset location, which type of account do we put the various assets. And so there are a lot of decisions. It isn't just how do I split up the portfolio, into which investment vehicles. I also find it valuable to understand what we're investing in, even if it's automated, even if we're just going to put it in the Vanguard Life Strategy Growth Fund, and there's nothing wrong with that, but we ought to recognize, well, what's, what's the yield of the bond component in terms of interest rate, the yield to maturity and am I comfortable with that yield when given the amount of interest rate risk, the potential for the bonds to fall if interest rates rise, Am I comfortable with how much is in US Stocks given valuations, while we can automate it, we don't want to do it naively. So we want to understand what current yields are, understand valuations, have reasonable expectations for a portfolio. So come up with a reasonable return and range of return for the Vanguard Life Strategy Growth Fund and run it through a model to see, okay, if this is the expected return and I continue to save this much and maybe do a simulation to see, well, what is the potential end of period wealth and could I live on that. That's a useful exercise and the kind of exercise that I think we should all do and that we would walk through helping 10 or 20 individuals do it in some type of live cohort. Now, this individual had a really diversified portfolio, maybe too, too diversified, but not really. Was 30% US stocks total stock market, 10% US small cap value, had 20% in international, so non US developed and emerging, 7.5% in international small cap value, 10% in TIPS, 5% in long term treasuries, 5% in intermediate municipal bonds, 5% in equity REITs, 5% in gold and 2.5% in Bitcoin. That portfolio is not dissimilar from others, but it does have more moving pieces. As you have more moving pieces, that means some pieces are doing worse than others. And it could lead one to want to micromanage and think, well, maybe that shouldn't be in there because it's not doing well. So that is the advantage of something like vasgx, because you're not looking at the pieces. You've essentially allocated that to Vanguard and that's perfectly fine again, as long as we know what a reasonable rate of return is for that fund and look at these other pieces that will impact the overall goal, which is our cumulative wealth or our total wealth after tax when we're ready to retire or if we are retired, will that wealth last? So I took the portfolio though, because I thought it was an interesting portfolio and I put it in Y charts. So I created a model portfolio out of it because I was interested to see how it returned relative to the Vanguard Life Growth Strategy Fund. And the model that it put together, it was rebalanced quarterly. It capped those target weights that he outlined. And because Bitcoin's fairly new, it's only a 2 1/2% weight, but it did have an impact. I could go back to May 2015, and his portfolio, if he had kept that allocation and started it over 10 years ago, would have returned 11.3% annualized versus 8.8% for the Vanguard Life Strategy Growth Fund. So about two and a half percent additional return. That would have led to much greater wealth. And the diversified portfolio was less risky in terms of the maximum drawdown. So the worse it return was 27.7% versus 28.5% for VAGSX. Now that that's not a big difference. In fact, the volatility was about the same. Each was about 13%. But because the model that he put together had a higher return, its Sharpe ratio, its return per unit of risk was almost 0.9, or it was 0.85 versus 0.69 for the Vanguard Life Strategy Growth Fund. But it did outperform. Now, Bitcoin made a difference even at two and a half percent, which goes back to something we've talked about a few years ago, taking aspirational risk. This was just an allocation. I don't know when he started allocating to Bitcoin. I started allocating to Bitcoin around 2016 or so, and it had a major impact in terms of my overall portfolio, even as I've rebalanced and I've kept it a fairly small allocation, although it has grown. But I thought, okay, well, what if we go back further? And by the way, that portfolio outperformed over pretty much every period going back to 2015. But if we swap out and say, okay, no Bitcoin, we'll just take that allocation and it'll have 7 and a half percent in gold and no Bitcoin at all. In this case, we can go back to April 2008 and it's still outperformed the diversified portfolio, but not by as much 8.2% versus 7.8% for the Vanguard Life Strategy Growth Fund. VAGSX had a higher maximum drawdown, so 51% return at one point. So I lost half the money versus 44% for this listener's portfolio. Volatility was, was about the same. But when you run a more diversified portfolio and you have these allocations, you always have the opportunity to allocate to the areas that have underperformed, perhaps have more attractive valuations within the bond portfolio. I know in our model portfolios we've benefited from having a higher yield to maturity than the Vanguard funds. And so the fixed income portfolio or models have done better than Vanguard funds. Not because we're smarter, it's just that we embedded a greater yield to maturity and less interest rate risk. And so we didn't lose as much money in 2022. There, there's a spectrum of how active we want to be as we save and invest for retirement. And there isn't an optimal answer. We're not optimizing, we're coping is what we're doing. And we're, we are making good enough decisions. But even a more diversified portfolio, it's good enough. It's not optimized. We can't optimize. In this book I'm writing, there's a metaphor I've been using called a capital reservoir. And it comes from a quote by Henry David Thoreau where he's talking about he's a walker, that his essay on walking. And he said that leisure and freedom and independence are the capital of the walking profession. Our capital is usually, we could think of it as financial assets, but it's much broader than that. And so we all have what I call a reservoir of capital, these assets. And it does include financial assets, but it also includes our freedom and our independence as our skills, our education, our health, our mobility, our time, our experiences, our relationships. It's everything that we've been gifted and accumulated during our lifetime. And it's tangible, it's intangible. And as our capital grows, our capacity to live more abundantly expands. And this capital is sort of our stock, it's this reservoir. And then we have flows into and out of it as we spend money, as we learn new things. But it isn't something that we can optimize. There isn't a right amount. We can calibrate it, we can add things to it, we can make trade offs to it. This gentleman owns gold in his portfolio. It's an allocation to protect against monetary debasement. It's possible to own gold and Bitcoin and not go all in and abandon the monetary system. So we're living right now where there's a lot of concern about AI. There's a concern about overbuilding of data centers and the environmental impact of that, the impact on utility cost. There's some pushback, there's some worry that there's too much investment and there'll be too much capacity and that'll hurt some of these hyperscalers or stocks. In fact, some of that concern has led us large cap growth stocks to underperform most areas of the global stock market over the past six months. And I have those concerns. At the same time, I still use AI. And we should use AI because it helps us expand our capacity. It increases our capital because we have more skills and ideally we're better learners. I wanted a book that I could talk to and discuss the contents of the book with. In fact, I'll seriously consider this next book. Maybe. I'll certainly write the book. But do we have an AI version of it that you can query the book? So that's where I stand right now. We got terrible experiences with AI, like with the M1 client service bot. There are aspects of AI I use every day. I recognize its flaws. It doesn't have perfect memory. Sometimes it has recency bias. There are elements of AI that I wish existed that don't. But we continue to experiment and recognize that we can't optimize our life. Good enough. Even when it comes to our investment portfolio. Good enough. Using rules of thumb. That's all we can do. We need slack in our life. We optimize everything. It becomes too rigid and too fragile. We need slack. We need things to be good enough. They can't be optimized. There's not a right answer to how to live a good Life. That's episode 552. Thanks for listening. Everything I've shared with you in this episode has been for general education. I'm not considered your risk situation. I'm not provided investment advice. This is simply general education on money investing in the economy. Have a great week. Sam. Sa.