Carter Malloy: Why better land data is the foundation of better housing supply

HousingWire

On this sponsored episode of Power House, Zeb Lowe sits down with Carter Malloy to discuss why better land data may be one of the most important and overlooked pieces of solving the housing supply crisis.   Malloy explains how his experience in land investing exposed a major industry problem: builders, developers, and investors were still making massive decisions using fragmented systems, outdated information, and hyper-local knowledge. That challenge eventually led to the creation of Acres, a platform designed to bring transparency, speed, and intelligence to land acquisition and development.   The conversation explores how better visibility into zoning, infrastructure, ownership, and growth patterns can help bring housing to market more efficiently. Malloy also breaks down the difference between AI hype and AI that actually creates operational value, arguing that the real opportunity lies in combining high-quality geospatial data with intelligent systems capable of analyzing land-use variables at scale.   Related to the episode:   ⁠Zeb Lowe’s LinkedIn⁠  https://www.linkedin.com/in/zebulon-lowe-a02353a4/  Carter Malloy's LinkedIn  https://www.linkedin.com/in/carter-malloy/  Acres  https://www.acres.com/   The Power House podcast brings the biggest names in housing to answer hard-hitting questions about industry trends, operational and growth strategy, and leadership. Join HousingWire’s Zeb Lowe every Thursday morning for candid con

Transcript

00:00In this sponsored episode of Powerhouse, I'm joined by Carter Malloy, CEO and founder of Acres.

00:05We dig into how better land and housing data can reshape the way our industry builds,

00:09invests, and plans for the future. We'll talk about why transparency at the land level has

00:14been such a persistent blind spot, how to separate real AI innovation from the noise,

00:18and what it takes to turn high-quality data into truly actionable insight for home builders,

00:23capital markets, and technology leaders.

00:37Carter, thank you for joining me.

00:39Thank you, Zip.

00:40Yeah, so I want to talk to you a bit. You and I chatted up quite a bit before this

00:46recording started,

00:47so I feel like I know the most important things about you. We both play guitar, and we both have

00:53an affection for gypsy jazz guitar, which is a really small sample size of people.

01:00But I'd like to learn a little bit more about you. So before we dive in, can you walk me

01:04through

01:05your journey to where you are now at Acres?

01:09Yeah. Personally, I grew up here in Arkansas and grew up around land and farmland.

01:16My career was really around investing and then turning that into land investing.

01:20So skipping past me and talking about the business, which is more fun for me to talk about,

01:26we began, I began personally 15 or so years ago investing in land and then turned it into

01:33a business seven or eight years ago to go help others deploy capital into land.

01:38Was it land investment only or was it for construction?

01:44Okay.

01:44That's correct. So just thinking about land. Today, same as a business, we are land only.

01:50We are obsessed with that specific problem. And as a company, we were building this as land

01:55investing platform. Our whole shtick was generate alpha through data science, right? We were going

02:00to have some nerds on staff, some geospatial sort of PhD type of folks that help us to go and

02:07understand land at scale and find good deals. And we ultimately, we got to a pace where we were

02:12investing hundreds of millions of dollars into land. And we kept banging into this problem, which is

02:18despite being, you know, sort of self, we like to call ourselves land nerds. It was still really hard

02:25to actually do that. The data was bad. The data was desperate. We had eight different software systems

02:31and logins we were using. And so as we were building this tool, which today is known as Acres,

02:39we, we hoped someday to commercialize it. We began commercializing that.

02:43Can I ask you a question? Can I ask you a question about that? Why was the data so bad?

02:47Was it because

02:48the technology wasn't there? Or is it because there wasn't anyone really investing in it?

02:51Yes. All of the above. So the primary system that folks use is Esri or ArcGIS. These are amazing tools

03:00if you're a scientist or an engineer. If you're a business person like me and like most people who

03:05actually buy and sell land, they are unusable. Again, it's not an insult. It's just not the

03:11purpose those tools were built for. So a lot of companies over the years that have stitched

03:14together little point solutions, but there was no real encompassing solution for land in the U.S.

03:21Couple that, that's sort of the application layer. Couple that with lack, general lack of availability

03:26of data, whether that's good topo, good sewer lines, good zoning, et cetera, et cetera. So that's really

03:33what we, what we went after.

03:36Okay. And so I know Acres is, you know, a position around bringing transparency both to land and housing

03:42data. And I'm curious to know why that's been such a, a blind spot or persistent, persistent blind spot in

03:50the, on the industry. I think a lot of it's been, the business has been hyper local is, is one

03:56and two

03:57is look, uh, you know, the, that's the way, that's the way my, my daddy did it. You know, it

04:02was, it's

04:02sort of a, uh, a real issue for, for us. And especially as we were in the world of farming

04:07of, you know, folks

04:08get into a habit. Like I'm, I'm used to driving around in my truck. I'm used to going and buying

04:11land

04:11this way. And so, and frankly, that's, that's a business challenge for us, which is showing folks like,

04:17Hey, spend an hour on your computer and save a week on the road. Right. Uh, so that, that,

04:23that, uh, transparency is in this, which is in our, our mission as a business is what we're all about.

04:28So I, I genuinely believe that markets function more effectively and people, uh, the individuals

04:34participating in those markets have way better outcomes when the playing field is, is even by,

04:39by data and transparency.

04:41Was there, have you experienced, I mean, to me, this seems like a win-win kind of, uh, solution

04:46here, but have you experienced resistance to wanting to bring the sort of transparency,

04:53uh, and with, with data to the, to this industry?

04:57Yes. Uh, everyone thinks they're getting a deal and, and that's often the case, but,

05:03you know, there's usually a losing side as well that you may or may not be on. And that may

05:09be a

05:09small part of it. A bigger part is really, as we started rolling out our competitive indexes

05:14and unmasking what people own around the U.S. Uh, that is, has caused a lot of consternation

05:21and we, we've gotten some real pushback. And what I mean there is, right. You, you've seen

05:26the stories of, oh, Bill Gates owns all this farmland or, oh, this, uh, this person owns land

05:30around the U.S. Well, the folks who are writing those reports are using acres to, to do the research

05:35at this point. And, um, and so we, we took that step further and built indexes. So housing is a

05:41great

05:42example where you can now track, uh, well over 90% of the production builders in the U S every

05:48one of

05:48them, every one of their entities and LLCs and what they're buying and selling, uh, in near real

05:53time and acres. And not everybody likes that. Right. Right. How much of that I'm curious as,

05:59uh, about, cause the way that I, the way that I understand it, you'll have larger builders that

06:03depending on the state or the region, they'll buy under an LLC or something that ultimately rolls

06:07up into the larger organization. And, uh, you, you know, if you're a smaller, if you're a smaller

06:13builder, uh, are you, you know, you want to build, uh, or high end homes, you may not be interested

06:18in

06:19buying a, you know, land that is 50, a hundred yards from, um, I guess, you know, uh, a rental

06:25unit,

06:26right. Or an apartment building. And I, I, how much of that is it, it's just the way that the

06:32industry has

06:33evolved or it's not, not arbitrarized isn't the right, isn't the right word, but there is a level

06:39of, of, uh, uh, I guess like a secrecy, uh, at play where, you know, other there's, there's a very

06:46specific reason why some builders do not want other builders to know where they're building and, and,

06:49and, and relative location to them. Does that, does that make sense? I don't know if I've phrased that

06:54correctly. It does. And I think ultimately, you know, I'll, I'll go to the most extreme examples,

06:59uh, which we, we track data centers, right. And, and for, you know, very real pushback,

07:04some of them are getting, um, in, in, in a lot of city councils around the U S. So we

07:08track the

07:08sentiment of the city councils for building and rezoning for home builders, data centers, uh,

07:13et cetera, um, through, through a company or a partner of ours, Hamlet, but the, the, the idea of,

07:19of tracking where companies are going, like you can see the same thing in retail. We help you

07:23understand where your competitors are going and they, they want to keep that under wraps. The

07:44the, the, the, the, the, the, the reality is, is the, the extreme majority of the time is people just

07:51view that as, as, um, you know, that that's mine, right. You don't, you don't get to know that.

07:55And, uh, I, I disagree simply. I think again, if we all have more information, we can all be more

08:03rational actors within the market. Right. Right. There's the, I mean, over the past,

08:09well, I mean, ever since the, the, the great rate rise post post COVID, uh, there's been so much

08:16conversation about, you know, rising rates and, um, and affordability and just inflation. I mean,

08:22a myriad of, uh, factors at play, but if you sit down and, and really it's, you know, in, in

08:28any

08:28industry that's tied, it's under the housing industry umbrella, it's, it gets to, uh, inventory,

08:36right? Like we have a housing, housing shortage in this country and the more homes that we can get

08:41out into the market, everything, all the other variables that people kind of are upset about or

08:46are attacking or kind of railing on, they, it's not like they're not important, but they become less

08:51important. And so inventory is kind of like the, the, the bottom line of, uh, of this, of every

08:58conversation. And, you know, I think that, um, the earliest stages, or it seems obvious to me that

09:05the earliest stages of, uh, of, of like land acquisition tie into, um, getting more homes on,

09:11on the, on the market and having more inventory. Can you kind of like, can you share or walk me

09:17through

09:17how you think, uh, industry leaders should be, begin kind of rethinking or reassessing that land

09:23acquisition as its overall place in the, in the formula of more inventory, more affordable housing?

09:30I believe on this theme of transparency is again, the more we know, the better, the better you can

09:35place, right? So where you're doing a deal, uh, when you're doing a deal and what you're pricing that

09:43deal at, uh, the more information you have, the more accurate you can be in, in assessing those,

09:48those, the, the variables there, uh, so that you have the best outcomes. And so not stacking on top

09:54of one another, like, you know, we, we want more housing supply. We also don't want, uh, homebuilders

09:59to have a bad business model, right? It's important that these folks can, can all deliver, uh, appropriately

10:03around, around the U S and all the right pressure points. So that's where, where we think, you know,

10:08measuring, uh, or sorry, uh, layering all that data, right? So understanding where the income

10:14growth is happening, understanding where the new plants and manufacturing is going, where the new

10:18job growth is headed, where the growth corridors are, where that new grocery store is, is going

10:23before the market knows that having all that data can help deliver more housing more effectively.

10:29Uh, one of the things I wanted to talk to you about, and you actually, you gave me a really

10:33cool,

10:33like very short, uh, demo, uh, was AI. And before I want to talk to you about what we'd say

10:40genuine

10:40or real AI versus what you and I were talking, calling it bullshit AI. Um, one of the problems

10:46I think that we have in, in, in, in our industry is that AI is just this very generic term.

10:52There's

10:52not an agreed upon definition of what AI is specifically as it's tied to any one industry.

10:57So within this industry, how, how would you define what AI actually is within this industry?

11:03Yeah. I think the, the problem you name is across all industries, right?

11:07Yeah, of course.

11:08And every, every person you meet like me that does anything with data or software says at this

11:13point that they are AI powered, right? Or AI native, AI first. Uh, and, and so I think,

11:19uh, you know, first we can define that, right? Like AI for a long time has meant machine learning,

11:24like, which is, you know, really great statistics more or less, um, uh, and, and decision trees,

11:29things like that. Um, there's computer vision, which is looking at things and

11:33trying to understand that the application there, which is quite applicable in the world

11:36of satellite imagery as an example. And there's LLMs, which is, you know, the AI of the moment.

11:41And, and for obvious reasons, it is a big deal. Um, and, and so the application of those things,

11:48you know, to, to me as a, as a user, um, you know, should be seamless and should feel magical.

11:55Right. If, if, if, if the, if something is called AI, but it doesn't feel that way,

11:59and it's not delivering you insane value, uh, then, then that's probably all you need to know

12:05about working with, with that company.

12:08Can this, this sounds like a silly question. How can you describe that feeling? But what does

12:12it mean? Whenever you say, when you're interacting with something, that's the awkward sounding

12:17question, but describe that feeling. What does magical feel like when you're interacting with

12:21AI to you? Um, you know, these are things that are hard to put into words, but every person

12:25listening today has had that moment where they typed in a question and just took a breath,

12:30like had, had a gasp, like, Whoa, right. This, this robot just did something pretty magical for me.

12:36Yeah. And that's the moment we're looking for when, when we talk about AI, like, you know,

12:41every company, their developers are using it to, to build software. Cool. Uh, lots of companies

12:46using it to go gather data, to go help them deliver things faster. And even some companies have,

12:51you know, what I would call lazy rappers, right? Like there, there's an LLM that's involved in their

12:55product, but doesn't do anything for you. Right. Um, uh, I go type some things

13:00and it applies some filters. Okay. Yeah. Like I real application, a real application of AI.

13:06I wish I had a better way to describe that, but it should save you such an amazing amount of

13:13time.

13:13Right. Usually what it boils down to is that, Holy cow. I can't believe this was able to produce

13:19this result for me that quickly. I know you and I are going to come, uh, just by the nature

13:23of what

13:24we do for a living, uh, are going to come to this from two very different angles. But I want

13:27to ask,

13:27I'll give you, I want to give you my moment with AI. And, uh, if you, if you can recall

13:32yours,

13:33like the first time that you had that experience, like, Oh wow. Holy shit. This is like, this is

13:37really something. Um, so it was, uh, I was, I did a LinkedIn post and I was, I was making

13:44fun of people

13:46that use AI to just generate their posts, which are obviously like predictable, um, you know, AI,

13:52it's AI slot. And so I, I had, uh, I had chat generate, uh, a, uh, an AI post where

14:02it adopted

14:03the tone of Hunter S Thompson, like fear and loathing in Las Vegas, fear and loathing in the

14:07campaign trail. And my post was something along the lines of like, if you're going to use AI to

14:11generate your, you know, your social content, I, you know, I suggest, uh, you know, adopting a tone.

14:16I like Hunter S Thompson. So I, and it was this article written, it could, it could have been a

14:21page from fear and loathing in Las Vegas or chapter fear and loathing in Las Vegas. I posted it. And

14:26like, uh, and it was a, it was a joke. I mean, it was awesome. It was the, the content

14:32was perfect.

14:32And, uh, I tweaked, I had to tweak it just a little bit, but the, but nobody got,

14:37maybe these people don't get my sense of humor, but everybody, it was so people thought that I actually

14:41wrote the thing that I was, that was just done by AI. I was like, wow, this is really

14:46um, one, maybe I have a skewed sense of humor, but two, this is really,

14:50this is really something special. Like the ability to really, if you spend a little bit of time with

14:54it, adopt a person's, adopt a person's tone to where it really is quite indistinguishable.

15:01Um, but I would assume that a data person, you'd kind of come with that from a different angle.

15:05Do you remember like the first time that you had that, that kind of like, wow experience?

15:09And we're like, well, I got, we got to, we got to hop on this train.

15:12Yeah. You know, look for, for far too many years, if somebody, and, uh, if somebody said the word

15:19Bitcoin or token or AI to me, I immediately did not take them seriously. I dismissed them as a,

15:25as a non-serious human. And, you know, I remember like when, when GPT three, five came out the,

15:31the end of November 2022. And that was, that was definitely the holy shit moment of, okay,

15:37first of all, woe, I've been wrong. And, and, you know, the, the implications of this and how quickly

15:42four, uh, rolled behind that. It was a sort of the scary moment of, Hey, this is, this is moving

15:46so

15:46fast that I'm not sure we are, we are ready for it, uh, as people. So, uh, you know, and

15:53since then,

15:54you know, it's been pretty consistent. I think last year, Opus four or five came out from Claude,

15:59that was a big deal. And then quite frankly, what our developers showed up with, uh, three months ago,

16:04which we released today, funny enough, uh, that was the big moment for me. We, we've been trying to,

16:10uh, crack geospatial AI, which is a whole new set of challenges, right? We're no longer extrapolating

16:16ones and zeros and, and, uh, you know, sentences, basically we're trying to do math about, you know,

16:22a hundred thousand or a million features like lines or points and polygons bent around the globe,

16:28right? The LLMs fall over dead. And the moment earlier this year that they showed up and said,

16:33Hey, like we, we think we've done something here. It was, um, like a hair raising moment

16:38internally for all of us of, oh my gosh, like we, we've, we have found something really special.

16:44Yeah. I think that, uh, that term bears, I guess, uh, repeating and, and maybe, uh,

16:50re-explaining or, or, or stating again, uh, geospatial AI unpack again, please. Uh, what exactly,

16:57what exactly that is. Yeah. So when we say the word geospatial, we're just talking about,

17:02you know, maps, right? And you have to remember that maps are not flat. They are, they are,

17:06they're wrapped around the earth. And so it presents very real compute problems. Like

17:11even something as simple as calculating the acreage of a parcel, uh, when you take into account the,

17:17the varying geometries, uh, of, of any individual parcel. And so, uh, AI has a much more difficult

17:24time exploring those data sets. And so, uh, what, what you can't do today at all is, is have complex

17:33AI queries inside of any of the frontier, uh, geospatial models. So what, what we do is, is we've

17:40built, uh, you know, pretty incredible system of, of tools and tricks and fast, uh, uh, geospatial tools,

17:47right? And we then have a number of AI agents inside of that, that understand how to operate

17:54and have context and, and weightings or trainings in various parts of our compute stack, uh, so that

18:00they can go and, and complete operations for you as a user very quickly. That I, I went off on

18:06a tangent.

18:07Here's what it does. Hey, go find me all the parcels over 40 acres near sewer with low topography.

18:15They're developable without flood that are either zoned R2 or reasonable in these three counties.

18:21And it comes back and does all that work for you and highlights them on a map. Go show me

18:26the lot

18:26pricing for all the lots, both finished and raw had been sold near Atlanta in the last five years.

18:32It uses all of our data sets and proprietary data to deliver you that, uh, in a couple of minutes.

18:38So does that, that's what, yeah, and what you were showing me, the, the, just a brief demo that you're

18:43showing before was, was wild. Does that require in order to do that or to deliver that, does that

18:49require a relationship or an ongoing relationship between both like your, your programmers as well

18:54as your software developers along with geologists, or is this, is this all a computing at this age?

19:01Is it now all a computing issue?

19:04Well, we, we, we certainly have, um, you know, PhD and master, uh, you know, geos master's program,

19:11geospatial, uh, data scientists on, on staff. So you, the, the, the expertise there is, is very,

19:17very niche and, and very important, uh, to be able to, to build the tooling, uh, that you can then

19:23basically, you know, guide these robots through to go, go teach them how to operate. And you have to

19:28couple that with very real LLM. And we also use, I mentioned earlier, you know, computer vision and

19:34machine learning. We use those forms of AI as well. Uh, but you, you've got to have some real

19:38expertise to go do the trainings and the weightings. And, you know, in our, in our case, that, that means

19:43like actual compute, uh, and GPUs here in our office, kind of a room, you know, full of our, a

19:50couple of

19:50our robots, right? That have really stupid names. Um, but, but we use, uh, to, to be able to sort

19:56of stay

19:56there on the, on the bleeding edge as the stuff is moving so fast and the rules of engagement change

20:01so frequently. Right. Well, I know that, uh, I mean, it's, uh, pretty common sense that, you know,

20:07the, the, the, uh, the cleaner, I guess, the data or the higher quality of the data, you know, that,

20:13that's essentially it's a foundation, right. For any of your AI driven insights. So what do you,

20:20how do you separate, you know, beyond like the, the magical, like, wow moment, how do you separate

20:25this like actionable data and insights from, I guess, the noise, like how should, uh, decision

20:32makers or leaders, how should they go about evaluating, you know, tools that they're investing

20:36in that are really delivering value versus like, what's Luke said, like what's bullshit.

20:42Yeah. Look, like the, the, the best analyst in the world, if you give them really bad data,

20:47they're going to produce your really crappy spreadsheet. Right. And the, the same rules absolutely

20:53apply garbage in garbage out with, with AI, right? Like there, there's an adage of like,

20:57show me, uh, bad data and amazing AI and I'll tell you which one wins. Right. So I'm glad you

21:03called

21:03this out. And it is, we, we identify as a data company first and foremost, right? So that, that is,

21:08uh, what, what most of our team is focused on. Most of our investment to date has been all about

21:13is bringing together high quality or high resolution data, right? So breadth, meaning like how many counties

21:19or, or cities or states or, or, uh, places do we cover, uh, depth, right? So how, how deep does

21:26that

21:26information go? How much do we know about every parcel on the US? Uh, and then, and then a recency,

21:31right? That's the, the, the third incredibly, uh, component, uh, incredibly, uh, important component

21:36there, uh, to sort of wrap all that, that context for, for, for data quality and data accuracy.

21:42Uh, so that, that's what we spend most of our time on still today is pulling all that information

21:48together for our clients to interact with and, and now for the AI to interact with on their behalf.

21:53I don't, I was going to ask you about earlier in our conversation, you'd said that you had

21:57experienced, um, like resistance, right. Well, with, uh, bringing the transparency of data to market

22:05and that there is kind of a, like, this is how my granddaddy did it, drive around in a truck.

22:10Are you at a, are you at a point now to where most people that you're interacting with on and,

22:16and either showing this or, or trying to, you know, sell this, or as you're trying to expand

22:21where people, cause it's, you know, AI adoption is, it's, is different though.

22:25I was just reading an article just the other day and in, in, in home building, AI is used, uh,

22:31surprisingly more and adopted surprisingly more than, than I would assume is much more than in

22:35real estate and, and the mortgage, uh, industry. Are you at a point where you're with the people

22:40that you're interacting with potential customers that say, um, they, they get it like they,

22:45they're, they know that they need to become more, uh, not even tech friendly,

22:50but they need to get on the AI train, or do you still experience that hesitancy or skepticism?

22:55Yeah. Look, our, our, our customers are cross industry working with land. Right. And, and we

23:00still see a healthy amount of skepticism, but, um, you know, I'll, I'll speak to the home builder

23:04vertical. I just came back from, from ULI and, and, you know, I'm, I'm always thrilled to hear the,

23:10the various use cases people have found for AI inside of their businesses. I think when it, when

23:16it comes specifically to land, uh, I, I, we are seeing it, right. We show people and, you know,

23:23the, the responses are usually some four or five letter words and like, Hey, that, that was my job

23:28five years ago. Right. That's, and that's true about a lot of AI is it's not, um, a super positive

23:34thing for entry level jobs. Uh, but at this point, like it is a matter of you can adopt this

23:41and use

23:41it, uh, or you can be at a competitive disadvantage. Do you see this as a opportunity

23:47and which I don't know how, um, I have no idea how, uh, let's say guarded the, this industry is,

23:57or your industry is specifically, but do you see an opportunity or do you see this democratizing

24:01perhaps like land acquisition to where, uh, people that, you know, uh, would not be drawn to

24:08the industry prior to this sort of technology, you know, this sort of transparency of data,

24:15do you see it opening up the industry, bringing in new, uh, new, maybe new talent, I guess is what

24:20I'm trying to say. I do because it makes it easier, right. And that's, you know, as a, as a

24:27company, it's always a challenge for us. How do we make these tools really friendly,

24:30right? Like you, you open your screen, what do you do next? It should be blatantly obvious. And so,

24:35uh, we, we've worked pretty hard and think, think we're, you know, made a lot of progress to this

24:39idea of I can hop right in at this point and begin speaking to the robot and, and get a

24:44really good

24:45sense of what I should and shouldn't be doing. Now, importantly is it does not replace experience,

24:51right? Like, like as we all know, trust, but verify, right? Like there, there's some amazing output.

24:57And the same is true for, for what, what our AI produces, like the, you know, the, uh, zoning

25:02report you and I were going through earlier, the, the, the outputs are, are great. And it's

25:09going to save you a lot of time. It's going to get you to say no to deals much faster,

25:12but you're also probably not going to just go make a multimillion dollar decision based on a robot.

25:18There's very real human judgment necessary, uh, to, to move these things along.

25:23Right. Well, we skip, I'll go back to my first question. We kind of skipped over this,

25:26uh, or fast forwarded right to, to acres. Uh, how, what got you into land acquisition? Like

25:31your first deal? You just dropped about a piece of property. You know, that looks like a good,

25:36like a good place to buy. Like where, where did that come from?

25:39Oh, my dad. No, I learned it the way my daddy did it. Just like I said earlier. So I

25:44am,

25:44I am certainly one of, one of those people. Uh, so yeah, my, my, uh, dad has been farming my

25:49whole life.

25:50So that got me on the country, a lot of the kid and, uh, got me to appreciate a sense

25:56of place.

25:56And, uh, I want to land, you know, that's, that's the, that's the dream.

26:00Excellent. Yeah, for sure. Um, okay. Let me ask you. So looking ahead, um, where do you see,

26:06or where do you feel most optimistic about the, uh, increasing adoption of AI and the advancements

26:14of, uh, technology in this, in the housing ecosystem at large, where are you, where two

26:19part question, where are you most, what are you most optimistic about? And then what gives you,

26:24uh, what, what gives you pause or gets you some concerns that you think people may not be, uh,

26:31thinking through or acknowledging. Well, I'll maybe zoom out further. So

26:36last year, a couple of really opposing books came out on AI. One was called, uh, AI snake. Well,

26:41and it's just about LLMs and hitting, hitting their ceiling and have really bad, uh, uh, financial

26:47models, which is somewhat true, especially in the latter part. And the other was called,

26:52if anyone builds it, everyone dies. And you can imagine what that one was about.

26:56And I read a similar one, uh, it was called, uh, against the machine by Paul Kingsnorth. And it made

27:01me want to, uh, retreat to the, to the mountains. And I, I look, I think it is important in,

27:07you know, anything you do in life to, to try to get an understanding of the opposing

27:11viewpoints as wild as they may seem sometimes. And, and that worry on that side is, is so

27:17material. It's so existential that we, even if it is a tiny, tiny, tiny fraction of a percentage of

27:23a threat, like it's something we should all be taking very seriously. Uh, so, so look, there,

27:28there is the, that, um, concerning side of, of, of a lot of technologies being built today. Uh, just as

27:34there was in the forties when we started building things that could destroy cities.

27:38Uh, and, and inversely though, like the, the unlocks that come from this technology are

27:45absolutely insane, right? Even if the technology stops evolving and this is truly the, the best

27:50model we ever get, uh, the implications are, are, are huge. And as, as we, as companies and as

27:57individuals figure out how to use it better and better, I think in housing, look, that means we,

28:01we reduce the friction, we reduce losses, right? So, so we help folks do better deals and think

28:06about just about land, but, uh, although that implies like better outcomes for the consumers,

28:11better outcomes for the businesses there and a, and a better functioning market. So I'm, I'm

28:15really, really excited about what it does for the companies, what it does for individuals and,

28:20you know, giving them quality of life because they're not having to waste their time on bad deals.

28:24Like I, and this is, it's an exciting moment we're at as me as, as acres.com and, and for

28:30all of us as an industry.

28:31All right. All right. Thank you so much for joining me. I really appreciated our, uh, conversation.

28:36I have, we have, uh, music stuff to talk about offline.

28:39Yeah, we do, man. Thanks, Sam.

Category

Transcript

Comments

Recommended