In this episode, institutional sentiment has ticked up roughly 5% from what was called the worst market conditions in 20+ years, while real estate remains stuck in a buyer’s market that hasn’t bottomed since mid-2022. AI is compressing market cycles faster than prior eras, forcing operators to check assumptions on a week-by-week basis against a backdrop of record corporate earnings and all-time-low consumer sentiment.
Key Takeaways:
- Market Cycles Are Collapsing News cycles and shrinking attention spans have pulled market reaction windows from months to days, forcing faster strategic pivots across every industry.
- Real Estate Is the Outlier The buyer’s market has dragged on since mid-2022 with no bottom in sight, while equities shook off the Iran war within weeks on the back of Magnificent Seven earnings.
- Top 10% Funds Half of US Consumption Wealth concentration is propping up the broader economy, and since land is a luxury asset, the K-shaped dynamic doesn’t materially change the SLC buyer base.
- AI Is Becoming the New Corporate Moat Ramp (now $34B run rate) built Project Glass on Claude Code as a self-correcting internal tool and deliberately refuses to release it, signaling that private AI capability is the competitive advantage.
- SLC Chat Accuracy Milestone The bot hit 95% accuracy on nuanced underwriting prompts using Sonnet and effectively 100% using Opus 4.7, built solo with zero external development cost.
Listen to the full episode to see why managing your AI usage inside Anthropic’s 4.5-hour blocks is now a top-tier operational skill that directly determines your competitive ceiling.
(Podcast transcript below)
Welcome to Get Serious, where at Serious Land Capital, we have successfully funded over six and a quarter million dollars worth of land deals with industry leading 41 % operating margins. So today I wanted to give another update on just markets in general, combined with some additional AI thoughts. Again, like this is the topic du jour for like almost every industry right now. And it’s just, moving so quickly that
you even on a week by week basis, you almost have to check your assumptions and, you know, be willing to update lines of thinking just potentially quicker than we ever might’ve expected in, you know, past market cycles and so forth. So, you know, a few key points here and different branch points to go off to, you know.
Looking at this market situation here and you’ve seen in some of my writing and some of the more recent podcasts, especially over the past couple months here, there was real sentiment from institutional capital that this was like, especially once that Iran war started, the worst overall market conditions that had been seen in a couple of decades potentially.
you know, even longer than that. and like there wasn’t necessarily an end in sight again, like uncertainty, just all time high. saw the consumer sentiment all time low since the data started being recorded in the U S and like the early 1950s. so, you know, that, that type of sentiment and,
You know, those type of statements like didn’t come from, from anywhere. And we were seeing just, you know, wild volatility in the stock market, which, you know, tends to be the loudest signal. Sometimes just, you know, a lot of noise too. It’s hard to figure out what exactly is going on. but more recently, you know, again, some, connections deeper in the institutional space, you know, over the past week or two, they’re like, you know, market.
slightly better, maybe like 5 % sentiment better, which, you know, as it compares to, you know, maybe the worst market conditions in like the past, uh, um, you know, 20 years or longer. Um, you have to take that with,
you know, just some appreciation rather that, hey, things didn’t come out to bottom out even worse or trend in a, or, you know, just kind of stay in the tank here. But I think what this kind of speaks to, and I’ve seen, you know, some other, you know, trusted individuals or at least, you know, trusted on my end with all, you know, access to trends amounts of data and, historical trends and so forth.
you know, seeing that oftentimes wars or wartime situations like don’t have that much impact at least on US markets. Again, it’s like we’re largely insulated from actually actual kinetic warfare and, you know, material, you know, access to material goods and, you know, day to day necessities that our citizenry needs.
So even like some of the worst conflicts and you know, the the world wars and So forth like you know more often than not like the market tended to bounce back within a year or so in most cases and so we see this Iran war situation, which is like The the sentiment here from the markets and this was this was kind of similar to like the tariffs last year and it’s like almost identical, you know some key changes, but
you know, where we had these massive volatile market shifts and, and, uh, know, changes almost on a day, day by day basis. Um, before the markets kind of like threw up their hands and it like, you know, we don’t know what to do with the uncertainty. You know, is there a blockade? Um, you know, no blockade, uh, or, you know, uh, again, is the straight or her moves going to be open or not? Or, you know, is there a ceasefire?
Again, in particular for this Iran war situation, like, hey, we’ve tried pricing this in, we’re trying to make reactions on a day by day basis. And the market’s kind of like throwing up its hands like, hey, can we just get to the part where we can all continue to make more money? The corporate earnings, still like some all time highs, Dow S &P and so forth. And a lot of that driven by the Magnificent Seven and
AI hyperscalers and so forth. again, like some of these corporations are, they’re arguably the best run companies in the history of the world. Like they’re just such finely tuned machines for producing cashflow. That even in the midst of all this uncertainty, like the market just like, yeah, okay, all this is going on. But, you know, at the same time, these corporate earnings are just like, they’re, they’re undeniable. So like, what, are we kind of doing here?
Um, trying to be too reactive and like, let’s just shorten up the market cycle. And so I think that is just a, um, you know, a telling trend that’s just becoming more apparent really across all parts of society is just things, you know, the, the window for, I mean, okay, well, to make this even more apparent, I mean, it’s like the new cycle, right? Like, you know, in the past, it might take, you know,
Uh, yeah, I more, more a follower from like sports or something. Like sometimes it could take, Hey, you wouldn’t hear about a game until like the day after. And, know, people might report on it and, know, some podcasts that I might follow that are more like on the sports side, you know, they were, um, you know, my, might come out like a day or even two days later. And it was like, no problem. This was less than 20 years ago, but now it’s just so such an instant reaction culture that it’s like, Oh, if you’re not.
recording on this, you know, basically right after the game ends, sometimes like even as the game is ending. So you’re just getting the most up to date info possible that like you’re missing the boat on reporting anything or getting any type of audience share because, you know, people are just so fine tuned to that real time information that things just become old news. You know, with, you know, like just
stunning alacrity at this point. And so we’re starting to see that again in so many other parts of society here. mean, look at attention spans, right? Like just shorter and shorter overall. So this is starting to play out on like a more societal level and like the market it’s about as big as it gets, is the market just following our collective or individual attention shortening?
And, you know, following along there. So it’s like, yeah, the market’s just realizing, Hey, can just move through cycles even faster than ever before. Yeah. You hit little blips. mean, look what happened during COVID. There’s a lot of different situations here, but like there was a big hit. I mean, a recession that lasted, I don’t know, a month or two. And then it was just like off to the races again. And we’ve seen some of these hits here with the market and there definitely are more concerning trends. And I talked extensively about, you know, like housing affordability and
jobs and the impact of AI and so forth. And you know, the K shaped economy where, you know, the rich are on average getting richer and you know, middle class and below just have a smaller and smaller share of, you know, total wealth and, you know, income within the country. Like those are all still very long-term and really, you know, medium, would say even short-term concerning trends here.
that are noticeable cracks. But at the same time, it’s like, you know, the market still is a primary driver. And it’s like, if we’re going to take any, you know, solace in something still providing more positive signals for the economy in general, like
Yeah, it’s still going to be better than everything collapsing at once, right? Or having warning signs all across and like you’d still rather have the wealthier folks in the country being able to continue to drive the economy and, you know, drive consumerism to, you know, prop up most other businesses and kind of keep things humming, even if it might not be as long-term sustainable, especially if like the rich get, you know, hit.
more substantially and then you lose a lot of that, know, half, literally the top 10 % again are funding half of the purchasing power of or, you know, the active consumer activity within the US on a yearly basis here. So again, we just have to be mindful of that, but also in the land space, real estate in general, you know, land’s generally going to be a luxury item. So
You know, you’re generally going to be catering to the top 10 % anyway. So it doesn’t really change the equation too much. You still want that to be done, but still some cracks in the economy, of course. but that’s just kind of a trend that I’m noting here is like, anytime we, and we, might think we might be entering like a tougher cycle and so forth. Things are just moving even faster than, than expected real estate, a bit of a different story. And like, that’s the largest asset class here.
it’s almost flipped on this. mean, because this has just been a protracted down cycle across again, but most of the country since, know, going since mid 2022, like it has just been a nastier buyers market that does not seem to bottomed out. I’ve written about and talked about a lot of those factors on why it’s been longer and harder for the
pain to play out across the economy or the real estate market rather. So we have to kind of keep all of that in mind, which just again can make it pretty frustrating when we’re trying to operate to the best of our abilities in this difficult real estate market. Whereas some of these other market movers, different industries, you the general stock market is moving so much quicker.
than what’s happening in real estate. So, some thoughts there, and this is clearly connected across multiple different tangents to AI. GN is just a key topic and something we’re spending all day working on.
Um, and yeah, there was all this talk about, you know, the AI bubble and the hyperscalers and, know, spending so much on the data centers and so forth, which again was like, you know, it’s it’s a key concern. Like there’s historical precedent for, um, know, such a significant chunk of GDP. Um, and like the biggest companies we have in the world, um, you know, let alone the U S you know, spending just incredible amounts of money, you
huge chunks of their cashflow, debt servicing, these AI build-outs when you were facing companies that might not have been deploying AI to the best of the bill, like is it really enhancing productivity and so forth? But you really see the past few months, especially as tools like Cloud Code and Cloud Cowork have come out that are just obvious to any company that has really started to
deploy and utilize them like, hey, this is the future of productivity here. Like this is the future of knowledge work to where these tools are just, you know, allowing solo builders to have, you know, the best time possible in their entire lives and just create, you know, these massive micro economies just almost overnight. You know, again, you’ve probably seen the starts with an I forget, but the
You know, the two brothers working on like now the, you know, $1.8 billion run rate, GLP one mailing service. That’s, know, pretty much fully run by AI here. like this is the potential that, so many people can, take advantage of here. And like the best companies that are the most AI forward are quickly realizing this and just spending gobs and gobs of money trying to gobble up as much compute and tokens from.
Uh, you know, these AI systems as possible with, you know, and, and, and, and, and, and, and, and, and, and,
yeah, my different type of companies here, but you know, revenue is revenue, right? yeah, we don’t know the exact profit margins. Some of the others are still, private, but, nevertheless, like it, it, it, is a point of comparison to, to note. and so again, like these companies were just, and we’re seeing it ourselves too, like, one of the most useful skills nowadays.
is just how you manage your AI usage because once you start applying this and building your own AI employees and schedule tasks and so forth, it becomes even more important to just throttle your AI usage and use the appropriate models so that you don’t get locked out of the system, especially Anthropic, where that’s pretty much the biggest concern, but they also have the best models to utilize. Because if you just burn out your usage,
you know, right now they’re like four and a half hour blocks and you blow out your usage, um, you know, within 30 minutes or an hour, uh, all of a sudden, like your system shut down and you can’t like do your best work for a handful of hours there. So like this is just a problem that all companies are trying to utilize AI is running into. these AI, you know, um, providers on the backend are trying to figure out, okay, how do we bump up capacity to meet this demand? Like, um,
It’s entirely possible where the conversation has shifted from, this is an AI bubble potentially, which could still exist. But, you know, people were worrying about, Hey, are we oversupplied here to where now it’s like, Hey, maybe the demand growing demand here. And like, I’m just seeing this everywhere where people are really getting the message, like, you know, credit to Kalan and her, you know, on common business team and,
Yeah, but they are just getting just insane amounts of signups for this latest iteration of their bootcamp and soon to be accelerated to automate. Like just logarithmic improvements, because I think people are just really seeing the potential of like the tools like coworking and so forth that are just more apparent for how they can fit within their business processes. And like also realizing, hey, you got to get on the train here because the haves and the have nots are going to separate really quickly.
and you see other companies like AI forward companies like ramp, you’re a newer company, but now they’re running like $34 billion a year or something. providing, you know, financial services to a lot of corporations, especially through credit cards. and you know, they more recently came out with this project glass, which is just this internal tool that, it’s almost. Well, it’s built on quad code, but it’s effectively.
um, you know, a self correcting software tool that anyone in the company can use regardless of technical capabilities, because it’s basically just a, you know, cowork type setup where you can just speak to it or, you know, type into it. And then it will just, you know, auto assist on what needs to be done or how does it update the code base? Like all of these things are just handled behind the scenes. Um, and also realizing, Hey, this needs to be adjusted here. So we’re making sure like the correct things are being deployed or not.
you know, based on the data that we’re getting in the company, hey, here’s, you know, probably a better product pathway to go over. So it’s just, you know, again, this internal system that, and it’s self-correcting, you know, lower cost comparatively to, know, traditionally much more expensive human employees and so forth and bumping the productivity of all the, human employees as well. And like they’ve deliberately said, Hey, they’re not releasing this public because they see their internal AI capabilities as a moat.
that they can continue to capitalize on, which I think you’re going to continue to see more and more of, whether it is just internal development of tools or just the capacity of companies to buy more AI usage and utilize the best models that the rest of the public doesn’t have access to. This is going to be the continued story, really, think, on a month-to-month basis here, and can continue to prop up the economy because
you know, what’s to stop like the best companies in the world from spending hundreds of thousands, millions of dollars a month, potentially for, the best possible models, the best possible compute by as much as they can. And again, these AI models just get becoming better almost on a week to week basis here. Like anytime I’m opening up Claude on the desktop app, like almost every day they have a new update. Okay. You got to reset because it is patch something and
you know, noticeable improvements, are, happening so quickly. So that to me is just another like market. You know, speed up mechanism, you know, the pace of AI is like going to filter into the pace of all other parts of society as well too. Like that’s kind of what it’s feeling like. So with that big of an impact on the market, yeah, that is the market kind of rebounding a bit too, because it’s realizing, like
There’s just no stopping this AI train from, you know, the demand that, continues to grow and the company’s taking advantage and like setting new precedent, like all these large, you better believe all the multi-billion million dollar companies, even smaller companies like ours, seeing what ramp has done with this internal tools. Like, okay, that’s the standard now. like the only way to compete is to build something similar, like that.
in order to stand a chance for where pretty much every industry is going. So it’s just a massive arms race that’s just happening so, so, so quickly. And again, like to note how good these actual models are here, like to share internally, mean, this SLC chat tool, which we’ve been working on for a few weeks, but even that sometimes has been piecemeal depending on other priorities. But when I’m testing like
tricky, tricky questions here. I’m just going to give you an example of, um, you know, a question that I asked this SLC chat, which, um, you know, if, if you weren’t as familiar, you heard me talk about or write about it before, you know, we’re just trying to build all our underwriting processes, everything we did for land price or into, you know, the serious land capital chat bot that really uses my voice, um, to answer any questions you might have about the land space, the general market, um, underwriting, comping, and so forth.
So like a test question we had was, you know, like walk me through what downside protection actually means on a specific deal. Subject is eight acres in a rural Southeast County asking price 48 K two recent sold comps at 7,200 breaker and 6,800 breaker, both similar access, flatter Topo. one older comp at 5,400 breaker. Show me the math. Like to me, I would say that’s like a pretty poor question because there’s so many details missing.
Like, you what are the characteristics of these accounts? Yeah, we know a little bit about the recent comes, but when did they sell anyway? You what was the days on market? How far are they from the subject? Like your own DD brain can probably start thinking about, Hey, this is like a tricky question for AI to answer, realistically. And, know, a lot of nuances required because so much is missing. but as I’ve iterated on that, like we basically got it to be.
95 % perfect with going through all the nuance and mathematical calculations using the mid tier. sonnet model for Claude and basically a hundred percent perfect using Opus 4.7. It is stunning. generally anything that might produce my voice and so forth, like it might usually get like 70 to 80 % there. But with the way we’re tuning this, that this bot here, like I basically don’t need to make any fixes for some of these questions. So.
And again, rapid nature of actually deploying this and developing it is it’s unreal. Like it’s just me to work working on it. Um, you know, not having to pay anybody else, like developing a full on like software project. Uh, again, we have all this underlying data, but like, you know, Claude is really doing a lot of the heavy lifting behind the scenes and producing results that are, you know, at worst 5 % off of how I would answer a question.
Um, that it again, it’s like pretty poorly worded and, um, uh, you know, trickier for AI to, generally answer because it’s missing so many details. Um, so just wanted to share that as like another example of just how fast things are moving and how much of an advantage it gives a company like ours. Um, the ability to have this both as an internal tool and, uh, you know, give to the general public because, um, uh,
you know, we, know our niche better than anybody else and we can apply the AI tools, um, to it. So, uh, that that’s the additional advantage here is like, okay, where can you pair AI to, um, you know, take advantage of unique knowledge that you already have. Um, so you can build tools that literally weren’t possible, um, or would cost a hundred X more even three or four months ago. Um, again, that’s the world we’re living in.
It’s getting faster and faster every single day. market cycles are getting faster. it was a brave new world out there. you have to adapt to it, erase all of your personal identities from before and just realize, Hey, you know, you might has to to strip everything down to figure out what your new position is going to be. when it comes to, surviving and, you know, ideally thriving as a business owner in this type of environment.
So hopefully this one was interesting and helpful for you all today. You can tell how pumped up I am about these subjects and our strategy of just pushing so many these AI tools here and also combining it to our real estate expertise and knowing, hey, we have a longer term pathway here. If we go heavier in the AI right now, because it just makes sense, get back to more,
deal pushes, especially if we make a lot more income from the AI side and can do a lot more larger deals anyway, especially maybe some cash flowing opportunities and so forth too, like the opportunities are infinite. that is our current approach at the moment, more to come soon. Subscribe and share everybody looking forward to next time.
