Skip to main content
Podcast

Neal Bawa — Operating a SMARTER Investment Portfolio With AI

In this episode of The Mailbox Money Show, host Bronson Hill and co-host Nate Hambrick sit down with multifamily data expert Neal Bawa to explore how leading operators are using AI to run smarter, more efficient portfolios—and what’s really happening in the multifamily market. They discuss why 49 of 50 U.S. markets are seeing positive occupancy growth, why finding the next “hot market” is getting harder, and how data is changing the way investors evaluate opportunities.

Neal also shares how his team uses AI, including a “mystery shopper” system that tests lead response times across properties, and why agentic AI could be the next major productivity breakthrough for real estate operators. If you’re investing in multifamily or want to understand how AI is reshaping real estate operations, this episode is packed with actionable insights.

YouTube

Neal Bawa is a multifamily data expert, technologist, and CEO of Grocapitus Investments and Financial Attunement. Known as the “Mad Scientist of Multifamily,” Neal combines real estate investing with advanced technology, data analytics, and automation to help investors make smarter decisions. He has been involved in thousands of real estate transactions and is a sought-after speaker and educator on topics including AI, data-driven investing, and the future of real estate operations.

Spotify

Get my new book: https://bronsonequity.com/fireyourself

Full Transcript:

Bronson Hill: Welcome to the Mailbox Money Show. I’m your host, Bronson Hill. We’ve got an amazing guest today. Before we jump into our guest, I want to introduce my co-host. We have Nate Hambrick, who’s the author of the 18 Laws of Leverage. Nate, let’s talk a little bit about how investing has been for you for the last couple of years. What changes have you seen in investing in particular?

Nate Hambrick: Well, sometimes it’s hard to tell if the investing landscape is changing or if I’m growing, and both are a good thing, right? The landscape is constantly gonna change. So, if you get into it and you think you learned everything, wait five years, right? But also, you grow as an investor because the more deals that you’re in, the more selective you are on the partners you decide to deal with, and the more narrow your focus becomes.

I know whenever I first got started, it was anything that would make money. And that’s not a bad place to start. So, if you’re listening today, that’s a great place to start. But the more information you know, the more selective you are, and the more narrow your focus becomes. And so that’s why I’m excited to chat with Neall again today, because he’s a master.

Hearing some of those honed niche insights will be very valuable.

Bronson Hill: Yeah, so that brings us to our guest today. We’ve got the amazing Neall Bawa here with us. I’m just gonna get into his bio a little bit. He’s a seasoned real estate investor, entrepreneur, and amateur data scientist. He’s known for his data-driven approach to investing. And he’s the CEO and founder of Grocapitus, which is a real estate investment company. They’ve acquired thousands of units. He’s the founder of Multifamily U. They teach investors how to leverage data, technology, and to make better investment decisions.

Neal, it’s great to see you. Thanks for being with us today.

Neal Bawa: Well, thanks for having me back on the podcast. I’m super excited to be back here. Seems to be a very interesting time to have conversations around multifamily and the economy.

Bronson Hill: Absolutely. So, I think you and I have some similarities. We did a lot of multifamily for a long time. And there was a saying, you know, we saw the rise in interest rates a few years ago. And then you brought it up before we started recording: was this “survive till 25,” right?

If you just survive till 25. Well, the pain, as you have mentioned, has lasted longer than we expected. And we’ve got our butt handed to us on some deals. Some we’ve done okay on, but it’s been challenging for a lot longer than people have thought.

And so again, with COVID, there was a lot of COVID money around, rates were super low, a lot of activity.

And things have changed. So, talk from your perspective, what you’re seeing right now. What’s changed over the last couple of years? Because it’s not 2025 or 2026. Like, what do we look forward to here?

Neal Bawa: Yeah, so I think the pain lasted longer for two reasons. There’s two major reasons for the pain. And the first one is supply, and the second one is interest rates. And if you basically, this is 95% of the pain if you talk about it. The remaining 5%, I would say, is investor sentiment.

But investor sentiment is fickle, and it changes if either one of these two things changes. Investor sentiment changes. So, those are the three things that are responsible for the pain.

And on the supply side, we believed that most of the supply glut that was going to come in would come in in 2024. And so by 2025, especially the second half of 2025, we would see that ebb. So, that was the conventional wisdom. The conventional wisdom was wrong, but it was the conventional wisdom of everyone, including myself. And what really happened is that the supply, instead of just coming in in 2024, came in in 2024 and 2025.

And now those units are still leasing up. Because if you have a 250-unit building that’s coming in in 2025, it’s still halfway through or two-thirds of the way through in the second half of 2026. So, that’s what we’re seeing.

We also saw that the Sunbelt markets appear to have two different waves of supply, not one. So, there was a wave that came in in 2024 and beginning of 2025, and then a wave that came in later in 2025 and the beginning of 2026.

So, some Sunbelt markets are still suffering. Austin’s a perfect example, San Antonio, Dallas. There’s these markets that still have this very large amount of inventory that is still coming in in 2026. And unfortunately, now that CoStar seems to be doing a better job of predicting this data.

When you look at CoStar reports for particular metros, they’re a bit sharper and more accurate than they were two or three years ago. And so we’re still seeing some metros that have another wave.

Coming in. So, for example, we have a property in San Marcos, and San Marcos has another 1,400 units that are predicted to be delivered in 2027. And that’s painful for us because we have a property in lease-up there and are having all sorts of trouble with it.

So, bottom line is that supply basically got spread out over time. And part of that was because interest rates went up so much that people that were building stuff had to raise more money. And in many cases, that meant that they had to either shut down construction or slow down construction while they raised more money.

And that basically added a year to the delivery time frames.

The good news is that, for the most part, with the exception of maybe six or seven major markets, most of the supply at this point in the second half of 2026 is done. So, we are now consistently beginning to see rent growth in most markets in the United States.

And we are seeing occupancy growth in almost every market in the United States, right? So, we may still be seeing negative rent growth in some of the markets, but 49 out of 50 markets now have positive occupancy growth where occupancies continue to move upwards.

That’s great. 49 out of 50 is a big number. I wish I could say that 49 out of 50 markets had positive rent growth, but that follows. It takes, you first basically suck up the occupancy, and as your occupancy gets into that 96, 97%—

That’s when people really start getting confident and getting rid of the concessions. Once you get rid of the concessions, that’s when the rent growth starts to be positive.

So, in the Sunbelt, the majority of metros at this point still have negative rent growth. But the projections from CoStar are that either some of those metros turn positive with rent growth in the second half of this year or in the first half of next year.

So, the supply story is the big change. That’s what’s really changing. And it’s changing month by month, quarter by quarter, market by market, but every single market that I’m aware of is making progress.

There is no market that I’m aware of, and I’m a little obsessed with this in an almost unhealthy fashion. So, I really do know the answer to this. If you look at major markets in the US, there is no market in the United States that is projected to have a higher rent loss than it was projected for the last 12 months. Always either a lower rent loss or a rent gain, right?

We’ve also now finally started to see some markets that have completely eliminated their supply. And now we’re beginning to see what happens when their supply is gone.

The first market in that list is Boise, Idaho, which now has completely exhausted all of its supply, and its rent growth on an annualized basis for the last quarter is at 6%.

Bronson Hill: ‘Cause that was super hot for a couple of years. That was like a few years ago. That was on fire, right? I mean, Idaho was, yeah. But now it’s coming back.

Neal Bawa: Yeah. Yep. Right. So, I mean, it’s a strong market, right? But in my mind, it’s not really necessarily any different from any other strong market. It just happens to be that Boise’s inventory was early, meaning its lenders stopped doing construction lending around 2023.

And because of that, most of its supply came in in 2024 and some of it came in in 2025. Here we are in the middle of 2026. They have no supply, right?

So, everything that the market has now is generic. When I say no supply, I mean just they have regular supply. Like, their supply looks right now like 2018 or 2019, right? So, they have some supply coming in. But when you look at the total absorption of the metro, that supply is deficient.

And so now we’re beginning to see 6% annualized rent growth. So, I’m hopeful, and again, this is not data-driven; this is just hopium, that we start seeing that in some other metros. Because if we start to see 10 or 15 metros follow the Boise path and start seeing 6% rent growth, then investor sentiment starts to turn around.

That’s what’s really needed for cap rates to turn around because they haven’t turned around at all.

Bronson Hill: Yeah. Let me ask you a question on this because I know you’ve, years ago, I remember seeing reports where you would always do, here’s the number one place for highest growth for this year or this city, and it was like, one of the Carolinas or was one other area, Utah, whatever. And it changes obviously every year.

Why is it now, with AI and with prediction models and some of the things that are out there, why can’t we better predict? Because it seems like what will happen is, like, my gosh, Boise is the best place, there’s whatever, and then all the developers now go to Boise and then you’ve got an oversupply.

So, it goes this constant, like, undersupply, oversupply, undersupply. Is there any way to make that more efficient, or is that just a market force and it is what it is and it will never actually get in front of that and kind of really be able to effectively know where development should be happening?

Neal Bawa: We can try getting in front of it, but it’s extraordinarily difficult because you basically include the listeners into that, right?

A market, you start seeing a lot of positive trends, but because data is widely available today and widely used today, now you have a very large number of people that become aware of the fact that this market, for example, is in a good place. Like, you know, land prices in Boise are not going down.

They’re going down in Texas and other places where land is almost free at this point in time.

But they’re not going down in Boise. Why? Because the Boise story is well known.

So, what we found actually is that when you look at the people that build stuff, especially it’s sort of the mid-market people. The top-of-the-market people, they always use data, right? Whether five years ago, 10 years ago, 20 years ago, they used data. But the mid-market people didn’t use data as well as they do today.

I think that AI makes it easier to use data. So, I think that there will never be a market that will continuously climb for five years again. I think that time has completely gone and cannot come back for any reasonably sized metro, simply because the data becomes available very quickly.

Nate Hambrick: So, how is that affecting your decision, knowing that you know so much about the data, knowing all the top-tier players and now the mid-level players and all the people with Claude and ChatGPT and Gemini are all getting access to some of the same data? How does that affect the decisions that you make, knowing other people are about to jump on a bandwagon?

Neal Bawa: Couple different things, right? You don’t anticipate a very significant boost, right? Your project has to underwrite based on it facing some challenges. So, there could be some challenges from some other wave of future supply.

The other thing is that you learn to time the supply. So, one of the things that I’m learning about is, imagine a supply curve, right? Supply goes up, goes down. That’s a curve. What I’m doing is I’m basically mapping the curve by metro. And I want to see—

Which are the metros where the curve has already crested? And then I’m looking at, and again, this is easier to do with AI. It was very hard to do before. So, what you do is you take 50 of these CoStar reports for 50 different metros, and you basically map the curve, and you say, which of these metros has the combination of two things?

Number one, the supply wave has clearly crested, kind of like Boise, right? And which of them— Are still not public, and there’s still not a sufficient number of permits, right?

Nationwide data about permits is really not useful. What it tells you is what the overall mindset of new construction in the US. So, useful for that purpose. But really, it’s localized permit data that’s useful, right?

So, imagine being able to take 50 markets and being able to compare permit data to that curve, and then basically asking AI to say—Which market has the widest opportunity gap?

It’s just a gap, and it’s a gap of one year or two years. After that, everyone will know the news, new construction will start, the gap will be filled, supply will come in. How big is the gap? How wide is the gap, right?

And can I take that gap, which supply is the biggest thing right now, and then slice and dice it against population growth, job growth, and income growth, the typical things that used to drive demand, right?

What I’m doing in my models is I’m giving supply basically an 80% weightage. And I’m giving everything else just a 20% weightage, and I’m trying to find the areas that basically have long-term prospects. By long term, I mean something that gives me two years to either buy or to build. Right? If it’s build, I really want three years, right?

And I think that I’m beginning to see some of those patterns there. But it’s not consistent. To be honest, in most places you don’t get three years. Why? Because there’s still a lot of supply.

I realize that multifamily permits have dropped. And one of the numbers that I always hear is it’s dropped 75% or 70% from peak. The problem is, why look at that peak? It was an irrational, crazy peak. We’re not gonna see it again, right?

If I look at how much it’s dropped from steady state, well, it’s only dropped by about 30 or 40 percent. So, there’s still a huge amount of new construction going on in this country.

And so it’s really hard to get a multi-year gap. So, Nate, I didn’t give you a good answer to the question, but the truth is that the availability of data means that gaps that are multi-year in nature in a great metro, not some metro that has negative population growth, are harder and harder to find.

In one way, data is actually making it harder for us to basically find these gaps.

So, I think the focus really needs to be on running the property better. So, I’m obsessed right now with using AI to run my property better. And I can give you guys some very fun examples of things that I’m doing at my property because I think that’s where we need to go, right? The magic is gone. And the next magic is, well, how do we run our properties better?

So, I’ll give you an example. There’s a company named U E F Y that makes security cameras. So, what we’re doing is we place two security cameras. They’re less than a hundred dollars, just so you know. Less than a hundred bucks each.

So, we put one camera in the leasing office, facing the door, and also having wherever the leasing manager is sitting visible. And then we have one camera directly outside the leasing office so you can see everyone that goes into the leasing office. And if the maintenance is a separate unit, then we put a third camera outside the maintenance room.

And we’ve written AI that basically has converted those because Eufy actually allows you to identify faces, right? So, we’ve identified the faces of every single person that works that property.

And the first thing that we do is we’ve built a timesheet. And so, what we’ve discovered is our properties, many of our properties, our staff was not consistently putting in eight hours a day. Okay? And basically, the property, the regional, also didn’t know that.

So, the regional property manager was surprised when we basically showed them an AI-generated timesheet. So, it says this person, on average, came in at this time and left at this time five days a week. And by the way, you notice that that’s only 35 hours, 36 hours.

And every time we do that, our regional property managers are surprised by that. Really? My staff isn’t working 40 hours a week, right?

So, that’s, and all of this is done through AI. I mean, basically, Eufy has an API, or application programming interface—

Bronson Hill: Yeah. Wow!

Neal Bawa: And that works for a hundred-dollar camera, right? So, we’re not building some fancy camera system. We are using Claude Code to write this application. And it’s had a dramatic impact.

Here’s a second application that we currently have. So, what we did was, for 12 properties, we created 24 Gmail email addresses. These are regular Gmail email addresses. Took a little bit of work to create them because Gmail, you know, you can only create a couple before it ties it to your phone number.

So, we did that. Created these 24 Gmail addresses and handed it to Cloud Code. And what Cloud Code does is that it fills out a form on each of our websites using that Gmail address, saying, you know, I’m a potential tenant, and it has the story and fills that out on the property website.

And then what we monitor is how quick is that property actually responding by email, text message, and phone.

Now, then the nice thing is the email—

Bronson Hill: I love it.

Neal Bawa: We usually get one email immediately and a text message immediately, but that’s an autoresponder. So, we just ignore it, right?

We are focused on what is real text messaging, real follow-up by email and by text message.

The one part that we have failed to build so far, because our AI is not that advanced, is that when they call, it’s always to leave a message. We don’t actually pick up because our AI would be, it would be obvious to them that they’re speaking to an AI, right?

So, when they call, they leave a voicemail. And so what we are doing is we’re analyzing the quality and quantity of incoming emails, text messages, and voicemails. Then ranking each of our property managers on a scale and then showing it to them on Monday morning, saying, this is where you rank.

You see, when we’re consistently looking at your follow-up. We don’t tell them what the email addresses are, and we rotate the same 24 email addresses across 12 properties, which means that it takes a while for that email address to come back, right? So, they never figure it out.

Nate Hambrick: This reminds me of Andrew Carnegie with the steel factories, where he used to put a plaque on top of the steel factory and tell everybody in the entire city exactly where that steel factory was ranking, and he got so much additional productivity out of it.

Bronson Hill: Make it competitive.

Neal Bawa: Yeah. It’s fantastic, right? I mean, just imagine basically looking at this beautifully designed page. It’s part of our MMRs now, right? So, we show it on Mondays, and it’s like, you are seventh when it comes to your text message follow-up. You are dead last when it comes to leaving voicemails compared to every one of my other properties, right?

And by the way, when it comes to asset living, you are third compared to two other asset-living properties that I have, right? Your local team—

Neal Bawa: Is way worse than the other two asset-managed asset living teams that are in other metros.

Imagine the amount of pressure that that puts on that property manager, right?

And the other thing is it’s a trend line, right? So, what we’re trying to say is, okay, we’re gonna shop you every single time. And these shops are expensive.

The reason that most of us don’t do secret shopping is Ellis is expensive. Ellis is a well-known secret shopping company that everyone in the industry uses. But the truth is, mid-market guys, we don’t have enough money to pay Ellis month after month. You’re not doing it, I’m not doing it.

But now AI is doing it, right? It’s not perfect because I still don’t have the ability to have a conversation with people because my AI, you know, is not smooth enough at this point in time to not be understood as AI. They’re gonna figure it out.

But five out of six things are done, and this app took me an afternoon to build. Well, it took me two afternoons to build, right? That’s powerful.

Bronson Hill: Yeah, it’s great.

Neal Bawa: Because the interface and design is awesome. They can exactly see where they’re ranked against every other property. And they can see a trend line.

Sometimes their trend line is getting worse. It’s like, wait a minute, the last time I shopped you, you were already pretty horrible. You were seven out of twelve. Now you’re nine out of twelve. What the fuck are you doing? Right? That’s the conversation that I’m having.

And imagine these two apps, and now imagine 20 other apps that I’m building, or what my company’s building.

Bronson Hill: It’s incredible. And now I think a couple takeaways, like Nate was saying, if you can create a kind of a level of competition and accountability, because the challenge is if you really don’t have that data, we say, I’m working, we’re just people aren’t coming in, whatever. And it’s like, well, no, they are coming in. Here’s what we get, we’re getting all this information we didn’t have before.

I did want to ask you one thing that I think is interesting, and you kind of brought this up a little bit too, is, like, you know, we can’t actually take those calls or turn those calls because people know that it’s AI. I actually want to know, like, what it, and you mentioned the term, is it real? Is this a real person?

And one thing I have noticed, Neil—

And I wanted to get your thoughts on this. Is, you know, last couple of years, it’s been harder to raise money. I had a deal. I was like, this is a great, it was a project we’re working on. I was like, I should have raised like one to two million for this. I’m hardly raising any money.

And then I just started picking up the phone and I called people, and it was me calling, not somebody on my team, but just me calling. And I had a couple of people I talked to that I realized they would not have invested had I not picked up the phone and called them.

So, it’s funny how, like, the more AI and all this stuff is going, people, we know it’s AI. And so, when someone actually does pick up the phone, it’s kind of going back to like the 1990s or 1980s when people actually picked up the phone.

So, talk about that. How’s this, like, the AI advances versus being a real person? How is that, ’cause eventually people are gonna always gonna want to connect with a person. But it may not be that, you know, certain times we don’t have to, and it’s just easier not to. So, can you talk about that dynamic a little bit?

Neal Bawa: Yeah. The answer is I haven’t found any kind of magic bullet or weapon when it comes to the investor side. Every example I just gave you is on the property side, right?

And I’ll give you one more example before I move over. I basically have an AI right now looking at every single multifamily insurance company in the United States and basically simultaneously sending packages, which it also creates itself, gorgeous-looking packages, to get insurance quotes.

Right. Before, we would simply not have the bandwidth to do it. Now I have one Claude agent. I have over 200 Claude agents now. One Claude agent whose only job it is is to constantly shop our properties.

Right. And it, yes, it annoys insurance agents, but why do I care? Right? My job is to protect my properties. So, we are constantly, we have one Claude agent just constantly shopping our properties to dozens and dozens of agents.

And most of the time we get the same answer, but every once in a while we get a different answer. And one property, different answer, $50,000 a year.

Magic, right? I mean, I’ll take it, right? 50 grand, it’s all NOI. It’s all NOI, right?

So, moving over to the other side. No, there is no way to replace hearing Bronson Hill’s voice, right? That can’t be replaced. And other people are doing it too. I haven’t gotten to that point yet, but I’ve been tempted to start calling my investors myself.

But I think the answer is— AI still has an incredibly large role to play when it comes to getting investors. It just doesn’t have a role to play on the phone, right?

So, let me tell you what has worked for us. So, we use an application called PhoneBurner. And PhoneBurner is an app that allows you to accelerate calling. So, I have a very experienced guy, his name is Peter Majeski. He’s worked for me for a long time. He’s investor relations director.

So, he’s the one making the phone calls, not me, right? What I’ve done is I’ve used AI to massively improve Peter’s velocity.

And how have I done that? Well, PhoneBurner has something known as an API, or an application programming interface key. So, I’ve gone in, I’ve grabbed the key, I’ve given it to Claude, and I said, I want you to analyze what the heck this guy Peter is doing.

First thing I realized is Peter’s not making enough phone calls. And then I go present that to Peter, and Peter cops to it and says, I’m sorry, you know, yeah, our volume fell down.

Then the next thing I basically told Claude is, I need you every single morning to have at least two hours of phone numbers for Peter uploaded into PhoneBurner.

Because the way PhoneBurner works is the magic of PhoneBurner is you click a button, you put on a headset, you click a button, and it just dials. And every time it hits a voicemail, it leaves the voicemail separately and dials a new call for you.

So, you’re just talking with people constantly. You’re never dealing with voicemails if you’ve configured the software properly.

Now, this is not an AI software. It’s existed for a long time. What’s changed is that now we have AI in charge of making sure that we are putting in the most nurtured people in our list into the phone numbers every day.

No human being is involved in that process.

So, here’s what happened. I send out some nurturing emails. You’re doing it all the time, Bronson, right? You send out some nurturing email.

All of the people that clicked on that email, they’re nurtured today. AI, so, you know, AI goes into ActiveCampaign, grabs those latest people, and adds them to the phone list for Peter to call the next day.

Why call people that are not picking up our emails? Why call people that are not engaged at the moment, right? I want to call those people, but I don’t want to use Peter’s time to do that. Does that make sense?

So, by doing that, by making that change, Peter’s—

You know, he always had a great tool, but now he’s making use of that tool in a better way than he was before.

So, we’re doing the same thing across LinkedIn. Like, for example, before you’d have a bunch of people that were listening to you on LinkedIn. I constantly get this email from LinkedIn every week saying, you know, 1,500 views for the week or 2,000 views for the week. 2,000 views is a good week, you know, 500 is a bad week.

Here’s the problem, though. Those people that are on LinkedIn—

Bronson Hill: Yeah.

Neal Bawa: You’re not making phone calls to them because you don’t have their phone number.

Now we’ve found a way to use AI and a third-party software to take everyone that’s connecting with us on LinkedIn today and flag it in our ActiveCampaign software so that their phone numbers are picked up.

Because two-thirds of those people are in our database. But they’re engaging us on LinkedIn. But when we’re making phone calls—

Nate Hambrick: So, I’m gonna hit you with a tough question here, Neil. Because I use Claude Code a lot. I’ve built multiple bots that do hundreds and hundreds, an incalculable amount of work that if you were to hire humans for, you just wouldn’t hire humans for because it’d be too expensive.

One of the challenges I’ve noticed is I’ve had a hard time explaining to somebody who’s never used Cowork, who’s never used Claude Code, how to get started. The best advice I’ve come up with was—

Nate Hambrick: Go to Claude Cowork and say, please install Claude Code and have it do it for you. And then just ask it questions until you find a use for it.

Do you have a better answer, though? So, for somebody who’s in real estate or just working a W-2 job, what advice would you give them to get started so that they can start developing agentic AI that actually does make a meaningful difference for them?

Neal Bawa: Watch videos. So, the absolute hack is this. There are a fantastic number of Claude Code videos that are on YouTube.

So, what you’re going to do is you’re going to go to Cowork and say, I want to learn how to use Claude Code as a vibe coder, but I am an impatient sort of guy. I want you to find me 20 videos on YouTube for me to watch.

And these 20 videos are people that are vibe coders like me that are creating applications.

Bronson Hill: Ha ha ha.

Neal Bawa: And they’re basically creating them pretty much from scratch, and the content is not too in-depth for me.

Now, first you go and watch those videos, and it’s gonna take you about an hour and a half, two hours to watch those 20 videos.

The next step is far more critical. You are then going to create your first application, and here’s that application: you’re going to tell Claude Code to watch five videos every week and build applications out of those videos.

So, you no longer need to give it instructions. So, I have a module, I call it “Web Brief, YouTube, Web Brief,” and it is in my mission control. And every single day it gives me five options, and I pick whatever I want to build out of it.

So, imagine not having to give instructions because somebody on the web is already doing it.

So, what’s the part that you need to bridge? You need to basically build one application which is—

Bronson Hill: Wow.

Neal Bawa: Go watch the freaking YouTube videos every day and tell me which ones we should be building for my purposes.

So, now you only have to describe your purpose. Whether your purpose is I want to get more investors, whether your purpose is I want to manage my properties, whether your purpose is I want to widen my funnel for acquisitions.

Explain the purpose. And then basically say, go find the YouTube videos that match this that already have instructions over—

Bronson Hill: Well, that’s amazing. So, I guess somebody who’s an investor who’s a past investor is listening to this, obviously on the operation side or the investor side of, like, hey, I look, I need more investors or want to operate my properties better. What’s the application for somebody who’s a physician or is a high-paid person or has a business?

Can you think of some ideas, just some examples? I know everybody’s business and occupation is different, but if I work for a company or I have a skill or I’m a trade or something, how would I be able to incorporate some of this into my…

Neal Bawa: You can. Anyone can incorporate it. I mean, the beauty of AI, one of the things I like to say is AI is the greatest invention of mankind since the invention of fire.

The other thing that I like to say, just so that you get how big a deal this is, if you take the internet, the smartphone revolution, and the personal computer, put them all together, multiply them a thousand, you still have not hit how revolutionary AI is.

The reason for that is AI fundamentally can make anything more efficient.

It’s just time that you have to put in. And the amount of time versus the productivity gains, it’s a spectacular amount, right? You can spend two hours working with AI once you’ve learned how to use it properly and you get massive gains.

In certain cases, I’m seeing 100x or 1000x, the use of my time coming back to me. So, this applies for anything. I think any business, any professional in the world, it applies for them.

Right now, for example, I’m building a better executive assistant. I call it Alice.

And so what I’ve done is I have a Philippines-based executive assistant. Her name is Michelle. So, what I’ve done is I’ve installed a software called WebWork, cost me $10 a month, on Alice, on Michelle’s computer.

The software takes a screenshot of Michelle’s computer 20 times a minute, right? Once every three seconds. And that means that in 24 hours, or, well, she works eight hours a day. So, it’s got hundreds, thousands of screenshots of what she’s doing, right?

At the end of the day, it tells me two things.

What is she doing that is manual that I can automate? And I have it write a full spec on automation, and I basically put it in the automation queue.

And then the second thing, the more obvious one is, what can you do that I don’t ever need to have Michelle do? So, you can replace it.

So, now it’s taking work that Michelle was doing. For example, all of my scheduling is no longer handled by Michelle. It’s handled by Alice, right? And it’s taken all of the scheduling over because that’s very easy for it to do, right? It’s something that can pick up very quickly.

By the end of the year, I may not have an executive assistant. Michelle is very bright, so we’ll just move her to another department in the company. But I don’t feel like I need an executive assistant.

And I’m somebody that’s had a Philippines-based executive assistant for a decade, and I still don’t feel like I need them anymore. Why? Because I feel like I am going to have an executive assistant. It’s just going to be called Alice, and it will do a better job than Michelle will.

And again, Michelle’s a wonderful assistant and knows my business better. But how does it stand up to Alice? Alice is listening to every Zoom call, every phone conversation, every Slack channel post, every Asana project that I have ever participated in, and building a second brain and using that to make decisions.

How does Michelle compete with that?

Bronson Hill: Yeah, it’s amazing. Yeah, it’s just interesting to see your evolution. You know, it’s like you’ve always been a data guy, but now it’s like with the data you take it and you give it to AI. You’re a guy—

Neal Bawa: I’m not a data guy anymore. I’m an AI guy. So, you can just call me that and say, Neil is no longer a data guy. And that is a perfectly valid statement.

I’m an AI guy. I work on it. I’m obsessed with it. I’ve never worked harder in my life. I wake up at six in the morning, I check my 10 or 15 Claude strings, but usually more like 25 Claude strings. I give them updates, you know, I go do my work.

And then what I’ve changed is my workload used to be, you know, between, let’s say, 8 and 4 PM. I’d just be booked solid with meetings.

I have now cut 50% of that workload, and I have meeting, AI time, meeting, AI time, meeting, AI time. So, I basically, 30 minutes out of every 60 minutes of my entire calendar for the whole week is working with my AI, right?

50% of my work time is just working with AI. And here’s the weird thing, right? You might be thinking that’s a lot. No, I’m thinking it’s too little. It’s too little, right? I’m not getting enough done because that’s—

Bronson Hill: Amazing.

Neal Bawa: 50% of my time is more productive than the other 50% of my time, which is still needed. You still need to go to meetings and talk with other people.

But I think that what I’m getting done because I’ve committed to 20 of my 40 work hours being AI only is dramatically more productive.

And it’s really hard to explain this to people that are like, AI is just me asking questions and getting responses back.

No, agentic AI is about you making sure that AI takes a task or a project or a set of tasks and completes it and then does it autonomously and then improves it.

And then you take that cycle and apply it thousands of times across your company and across your personal life. I actually like employing it across my personal life better because you get all kinds of interesting ideas that way.

Bronson Hill: I love it. I got one more question for you, and it might have another question too, but this is a little more philosophical question on AI.

Dario, who leads Anthropic, said famously a year ago that new graduates would have a 50 to 60 percent unemployment in one to five years. And the other AI leaders, such as Sam Altman and whatever, have said it’s kind of the opposite.

We’re seeing there’s so much demand, there’s we can’t hire enough people. And so, a lot of people obviously are very afraid of AI.

But sometimes the promise of technology, because I see it as basically an extension of technology, it’s that it’s going to lead to less work. But I don’t know anybody’s working less because of AI here. Like, for you, I say I’m building all this stuff and I’m doing this. And I think it is the AI-empowered entrepreneur and employee and whatever is going to beat out the other companies that are not using it, right? It’s really, it is very important.

But do you think that there’s ever a point where I guess we don’t need to work in the same way, or that a lot of this is taken over, or is it just that we just continue to find new things to do and work more?

Neal Bawa: I think they’re both right. It’s just hard to tell which one is right first, right?

So, the short answer is this. I do expect at some point the 40-hour work week to become a 32-hour work week and then become even shorter than that. And this has happened before.

On average, Americans worked about 60-hour weeks a hundred years ago, right? So, today we are working about 37 or 38 hours. That’s the average number. Europeans work about 33 hours. So, we work about five hours a week more than Europeans do. And the reason for that is productivity, right?

So, in the next five years, we are going to see the greatest productivity boom that we have ever seen in history.

That productivity boom allows, for example, for us to keep growing our debt. You know, one of the best ways to keep growing your debt, the fiscal debt, is to have five percent productivity every year. If you do that, by the way, that debt shrinks as a percentage of your economy, it shrinks. In dollars, it’s going to keep going up.

So, I’m actually much more bullish about the federal debt than I was five years ago, where I thought that we were completely doomed. Now I’m not so worried about it.

I do think that we are going to see significant spikes of unemployment amongst younger folks that are coming out of schools and colleges. I do not think that the overall economy is going to see unemployment rates.

Companies are hiring back the people that they laid off for AI. That’s a trend that we’ve seen in the last quarter. So, I’m not overly worried about where the world economy and the US economy are going because of AI, and in general worried about the fact that we seem to be fighting ever more wars.

But related to AI, I’m not massively worried about it. Because I think what AI is about to do is to unleash an astonishing gain in productivity, and that helps everyone on this planet.

Productivity gains, people don’t talk about them often—

Are a magic weapon that we have to grow the size of the world economy. And we are about to get that.

And you might say AI has been here for four and a half years. Well, not quite four and a half years, three and a half years, right? So, November 30th, 2022 is when ChatGPT came out. So, we’ve had about three and a half years.

My argument, though, is agentic AI has only been here for seven months. To me, agentic AI is a hundred times more powerful than AI.

And agentic AIs came in with OpenClaw, right? And then with Claude Code and now a whole bunch of things, a whole bunch of other things have come in.

And there’s only very few people that fully understand how different agentic AI is from regular AI. And most people don’t understand it.

I myself am struggling with it because it’s so shockingly powerful that I struggle with it, even though I use it 12 hours a day. And I think maybe—

Less than one tenth of one percent of the world population understands the benefits of agentic AI fully. And when we get to the point where ten percent of the world population understands it, we will see extraordinary gains in productivity.

So, all of the productivity gains of AI, all of them are ahead of us. And that’s what gives me hope that we are actually going to land this thing and land it well.

Bronson Hill: Love it. Well, Neal, I just wanna appreciate you, all the value you bring over the years, both the data and, like you said, AI, and just really staying on the front of this and really creating in the space.

How can people connect with you and reach out to you if they wanna connect and learn more about your deals and your education and what you’re working on?

Neal Bawa: Sure. So, the best place to start if you’re interested in what we are doing with AI, especially the real estate side of AI, right? People are interested in, Neil, what are you giving away?

Well, I’m giving away completely built applications. Take it and we’ll give you a one-single-line installation. Go install it. It’s an astonishing application. We’re giving those kinds of things away every month.

That is grocapitus.com. That’s my company, G R O C A P I T U S dot com slash agent, agentic AI, agent.

And if you’re interested in all of our real estate data, we publish a massive amount of data, eight webinars a year; about 10,000 people sign up for those.

That is at Multifamily University. You can Google it, or multifamilyu.com/club. It’s completely free. We don’t have any paid education. We don’t have, at this point, an intention of creating paid education for real estate.

But you might want to check that out. Multifamilyu.com/club, where we publish all of our data, all of our—

Bronson Hill: Awesome. Thanks, Neil. I appreciate you, man. Thanks for being here. We’ll just stick around for a minute. We’ll break down the episode here.

Nate, a lot from this episode, man. I feel like I learned a ton, and obviously I talked to Neal, I always learn something. But what are some takeaways for you on some of the AI stuff we talked about or the investing?

Nate Hambrick: Well, I’ve got something better than a takeaway. I’ve got an application. And I, Nate Hambrick, living in Dallas, Texas, will be creating one agent that watches five videos with instructions. So, I don’t need to give it any prompts to figure out applications for my business.

I think that is a stroke of genius. I wish I came up with it myself. So, thank you, Neil. How about you, Bronson?

Bronson Hill: Yeah, I think just the focus on continuing to find improvements. I think, you know, I’ve kind of been, I don’t say I’ve been avoiding Claude Code, but it feels like I don’t know if I’m a coder or whatever, but just getting more, taking more effort in it.

‘Cause as people I’ve seen that really get into it, they’re able to really generate a lot of value. And I’ve seen it, the things I have used it for, it has created a lot of value for me.

So, I think I’ve gotten a lot of value just from being willing to try new things, being willing to spend time doing it. I loved the takeaway of, hey, I used to spend all this time doing meetings, now I do 50% AI and I want to do more AI because it’s just, I’m getting so much out of it.

So, that was pretty powerful. So, I think that was great.

Well, just want to say to our audience, thank you again for being here. So grateful that you choose to join us.

If you haven’t joined our investment club, you can check out the link below or go to bronsonequity.com. We’re doing some exciting new projects coming up. We’d love to partner with you on some of those.

And really, the best thing you can do is to be in rooms like this where you’re educating yourself. You’ve been spending time with Bronson Hill, Neal Bawa, and Nate Hambrick.

So, we’ll look forward to seeing you on the next episode of the Mailbox Money Show. Thanks, everyone.

Outro: You’ve been listening to the Mailbox Money podcast.

For more free resources articles and videos go to bronsonequity.com. There you can download your copy of The Special Report on the Single Best Investment Strategy During and After a Pandemic. None of the information shared here is an offer to buy a specific investment, and this is for educational purposes only.

Consult your financial legal and tax professionals and use your own common sense before making any investment decisions.

Thanks for joining us and be sure to tune in next time for more Mailbox Money!

 
Bronson Hill

Bronson used to work as a consultant for a medical device company but switched to investing in apartment buildings to make his money work for him. He started with a single rental property that made good money and, after some advice from a family member, moved into bigger real estate projects. Now, he's all about helping others get into this kind of investment to earn money without having to work all the time. When he's not dealing with investments, Bronson loves to travel, write songs, stay active, and help fight modern slavery through his work with Dressember. He believes in working smarter, not harder, and wants to share how that's possible with everyone.

Leave a Reply

Are you human? Please solve:Captcha