Latest / TFTC: A Bitcoin Podcast / #752: Why AI Stocks Are Cheap with John Tinsman

Transcript

#752: Why AI Stocks Are Cheap with John Tinsman · TFTC: A Bitcoin Podcast

• 0:00
  John Tinsman, welcome to the show.
• 0:02
  It's a pleasure to be here, Marty.
• 0:03
  Thank you.
• 0:04
  It's a pleasure to have you on such short notice, too.
• 0:06
  I think we threw this together less than 12 hours ago.
• 0:09
  I'm excited to have this conversation because we've had analysts covering the AI space on
• 0:15
  quite a few times over the last year.
• 0:17
  Jordy Visser, probably the most prominent.
• 0:19
  I've really liked his coverage and was telling you before we hit record, I found some clips
• 0:25
  of you talking about your thesis and what you're doing and said I had to reach out to
• 0:29
  John, get him on the show. And for anybody who's unaware, John is the founder and portfolio manager
• 0:35
  of AOT Invest, which trades on NASDAQ. He's also the director of procurement and investments at
• 0:41
  Twin State Inc. Family fertilizer business, which is maybe a topic we can get to later. But I think
• 0:46
  really respecting your time and jumping into it. I think first laying out your framework,
• 0:54
  I believe your thesis is low marginal cost, high growth businesses trying to find who can outlay the least capital and get the highest multiple on that capital for your ETFs.
• 1:08
  And I think just walking us through that, because most people hear AI investing, they think NVIDIA, but your framework filters for something more specific from what I understand.
• 1:17
  Yeah. So my first ETF I launched in 2022 is AOTG.
• 1:21
  That's the ticker, the AOT Growth and Innovation ETF.
• 1:25
  So I came from the world of market making, which was all about high speed and algorithmic trading and statistical probability.
• 1:35
  So I tried to take that statistical probability to long term investing.
• 1:40
  And so when you invest 10 years from now, you can almost guarantee that the stocks that have gone up the most have had the best earnings growth.
• 1:48
  So like in 10 years, if there's a company that's earning 30x more, that is probably have gone up quite a bit.
• 1:54
  And so I think like when you invest for the future, you think, well, what companies give me the best probability of earning the most in the future?
• 2:02
  And so the first thing I came back to was the companies that are earning the most today, like growing the fastest earnings today.
• 2:07
  So like companies that have really high earnings growth statistically are the most likely to be doing that again tomorrow.
• 2:12
  So I target and I wait by high earnings growth and high revenue growth.
• 2:17
  which is really unlike modern investments, I think, because the modern investments,
• 2:22
  I think, fall into two categories. You've got the index funds that are waiting by market cap.
• 2:27
  So if you have a growth fund, Apple might be a top weight, and Apple's barely growing faster than
• 2:32
  inflation. And so why would you not put a top weight in something like Micron instead if they're
• 2:38
  growing their revenue at 300% year over year? It's a good question to ask. Why would you not
• 2:43
  weight higher on companies like AMD and SanDisk if they're growing their earnings at 1,000% year
• 2:50
  over year. You would think in a growth fund, you would want the highest possible growth.
• 2:54
  So that's what I do in AOTG. And then the other thing about active management,
• 3:00
  modern active management, is that they focus a lot on the betas and the sharp ratios and things like
• 3:06
  that. And that was counterintuitive because you're using a backward-looking volatility metric.
• 3:11
  So like, it's not necessarily indicative of the future.
• 3:14
  Meanwhile, you kind of avoid companies that are doing really well if they just go up and
• 3:18
  down too much, maybe like Sandisk and Micron.
• 3:20
  So like, I kind of take a common sense approach to buying high growth companies at the most
• 3:25
  reasonable valuation metrics possible.
• 3:27
  So if there's something like super overpriced, like Palantir, it won't be in my fund.
• 3:32
  And, you know, even if it is decent growth, like why would you even bother wasting your
• 3:36
  ammunition on it if you can buy something at one 10th evaluation metrics, three times
• 3:40
  the growth?
• 3:41
  So I take that approach with AOTG. And then number two, the second highest thing I found correlated with long term stock price performance in my research was low marginal cost.
• 3:53
  And I think this one's really interesting because nobody else was focusing on it. And that's the second criteria for AOTG, high growth and low marginal cost.
• 4:00
  So low marginal cost is the cost to produce one additional unit of good. So think about a company like Microsoft.
• 4:06
  If they make one additional unit of Microsoft Office and sell it to you, that costs them nothing to make.
• 4:13
  That means they have 100% profit margin, which is great for shareholders.
• 4:17
  That means they can scale to meet demand without any debt.
• 4:21
  And then they can also have an enormous amount of free cash flow to keep innovating.
• 4:25
  So you've got higher innovation, higher profits, higher growth.
• 4:29
  You compare that to a traditional company like Boeing.
• 4:31
  if they have to build another airplane for, you know, and they're selling it for a hundred million
• 4:36
  dollars, it might actually cost them $99 million. And if they really want to grow, so they have a
• 4:40
  very low profit margin on new sales. And if they really want to grow, they have to build new
• 4:45
  factories and then they have to sell product at a negative profit margin to steal the business
• 4:50
  from a company like Airbus. So growth, not only is it funded by huge amounts of debt, it comes
• 4:55
  instead with a hundred percent profit margins. It comes with negative profit margins and losses.
• 4:59
  So that's traditional industry. I learned this in my family business, a fertilizer,
• 5:04
  because I always like, I'm ambitious guys, like let's grow the business. And everybody's like,
• 5:08
  no, if you grow the business, you'll just light your money on fire and you'll never see it again.
• 5:12
  And so it's like, let's just be happy with what we have. And it was hard for me to wrap my head
• 5:16
  around that being, you know, wanting to grow and do better than the year before every year.
• 5:20
  Um, so, uh, and so I take that investment approach, uh, that I've learned from personal,
• 5:28
  you know, managing family businesses and took it to an ETF and focus on that high growth,
• 5:34
  low marginal costs within profitable businesses.
• 5:36
  And so it's been out for, you know, three and a half or four years and it's up, you know,
• 5:42
  it launched at 24, um, on the NASDAQ and it's trading.
• 5:46
  I think it traded yesterday in the 63.
• 5:48
  So in that short amount of time, it's up over 160%.
• 5:51
  So it's been great to do well for shareholders, of which I'm a big one, of course.
• 5:55
  It's like, oh, and I have all my money in.
• 5:58
  Well, I mean, jumping like deeper into the framework and particularly, I think your hottest
• 6:04
  thesis right now, which is being all over this XAI Anthropic deal, which was announced,
• 6:09
  I believe, last week or the week before, or Anthropic, I believe they're paying what?
• 6:12
  1.9, 1.6 to 1.9 billion a month in compute from one of the Colossus sites.
• 6:19
  Yeah.
• 6:19
  Yeah.
• 6:19
  And I tweeted about it because I hadn't seen anybody talk about it.
• 6:23
  I'd seen people talk about, oh, they're leasing it, but nobody had talked about the profitability
• 6:26
  metrics.
• 6:27
  They were open up the, they opened up, it's so crazy.
• 6:30
  Marty, they opened up the S1 and they're like, oh, ha ha ha.
• 6:33
  Look at how much money XAI is losing.
• 6:36
  And in that, they basically had $15 billion a year coming online like next month.
• 6:41
  So like in like three days and they somehow didn't realize that it's going to be making
• 6:45
  money.
• 6:46
  Like, not only is the XAI revenue going to jump 300%, maybe a lot more, but a year over year, but also they're going to be making a ton of money.
• 6:56
  And so, and so the, so like everybody was just kind of wrong on their take.
• 7:01
  And so I tweeted this and like all the celebrities are in the replies.
• 7:04
  It's gotten like hundreds of thousands of views on X, maybe millions by now I haven't checked.
• 7:08
  But, but the interesting thing is that they spent three to 4 billion.
• 7:14
  And some people say as high as $7 billion.
• 7:16
  And these are kind of closely guarded secrets.
• 7:18
  You kind of have to look at how much XAI was spending and try to say, okay, how much of that did they spend on Colossus 1?
• 7:23
  But they built a 300 megawatt data center.
• 7:26
  They did it in 122 days.
• 7:30
  And, you know, this thing probably costs a couple hundred million dollars a year to run on the very highest side.
• 7:34
  You know, it might cost only $150.
• 7:36
  I don't know.
• 7:37
  Um, but so if they spent 4 billion on it, let's just say, and they, they lease it to XAI for $15
• 7:44
  billion a year for three years, it will make $45 million on a $4 million invest a billion,
• 7:50
  $45 billion on a $4 billion investment. So that's a 10 X ROI. And I think in three years,
• 7:56
  the asset is probably will appreciate it in value. So like, I don't think like it's depreciating.
• 8:01
  I think it'll still be very useful in three years. So the ROI is really a lot higher than 10X,
• 8:08
  in my opinion. And so that's like the tip of the iceberg, because I think this is a closely
• 8:13
  guarded secret in the industry. Nobody wants to talk about how much money they're making in data
• 8:18
  centers because they don't want everybody to know. At the same time, the hyperscalers are spending a
• 8:23
  trillion dollars now this year building them. And did you know Microsoft and Google, historically,
• 8:28
  their return on capital invested is over 35%. And why are they spending so much more now?
• 8:34
  Is the ROI trending upwards to 100%? And if you look at Google Cloud, if you look at Microsoft
• 8:40
  Azure, if you look at Amazon Web Services, Oracle, it seems like it quite possibly is.
• 8:46
  The revenue growth is accelerating. The profit margins are getting fatter and fatter.
• 8:51
  And I think there's a chance that what we're doing is we're building data centers in the US
• 8:55
  and releasing it out to the rest of the world at like 90% profit margins.
• 9:00
  And I live in Iowa.
• 9:01
  I'm not sure where you are, Marty.
• 9:03
  But like the data center boom here is huge.
• 9:06
  Like my neighbors work for Oracle, electricians and welders that used to be making like $50,000
• 9:10
  a year, making $250,000 a year building these things.
• 9:14
  And like they're going up everywhere.
• 9:17
  And it's just bringing an enormous amount of wealth.
• 9:19
  But here's the crazy thing.
• 9:20
  You talk to these people and they're booked out five years.
• 9:23
  Like the data center demand is not going down.
• 9:27
  Meanwhile, Anthropic grew their tokens sold by 80X in 12 months.
• 9:34
  12 months.
• 9:35
  And Goldman Sachs is saying they think they're going to see another 24X token demand growth.
• 9:41
  And so we are so far past 20% revenue growth, 34% revenue growth.
• 9:46
  When you're starting 80X, 24X, how do you even value that?
• 9:51
  And it's so bullish.
• 9:52
  The ROI is so good on data centers that as people really going to order less semiconductors, because people are like, are we at the top of a cycle?
• 9:59
  And you look at the numbers and you're like, this is absurd.
• 10:01
  I've never seen anything like this before.
• 10:03
  And so I think there's like why it's always good when you invest to have a little bit of pessimism and make sure you keep in mind the downside potential.
• 10:11
  I think it's really good to be focused on the upside right now and not to kid yourself about like what the potential is.
• 10:17
  Can't be dooming. Can't be doing especially not right now.
• 10:19
  And it's funny today in our newsletter, we actually covered it.
• 10:21
  We've been sleuthing the banking reports.
• 10:24
  One came out of Morgan Stanley yesterday where they're citing these CapEx numbers.
• 10:28
  And they're warning, they're saying, has AI made the U.S. economy inelastic?
• 10:34
  Hyperscaler CapEx estimates nearly doubled to $805 billion for 2026, headed towards $1.1 trillion in 2027.
• 10:41
  U.S. GDP has been revised up to plus 2.3%.
• 10:44
  S&P earnings growth revised up plus 23%.
• 10:47
  33%. Their thesis is that Google, Microsoft, Meta, and Amazon aren't cutting CapEx in response to
• 10:52
  higher rates because AI spending is too strategic, too essential, or too well-funded to care.
• 10:57
  If the biggest spenders in the economy are rate insensitive, the Fed's transmission mechanism is
• 11:02
  weaker than at any point in history. So they're making a comment on this CapEx spending as
• 11:07
  juxtaposed to high rates and what the Fed may want to do. Moving forward, I think to your point,
• 11:14
  what people haven't really grokked is the ROI on these CapEx investments, particularly if you're,
• 11:20
  you own the compute and the energy infrastructure is so high, it's hard to pass up.
• 11:25
  Yeah. I mean, I think there's two things to think there. One, like if the Fed raises rates one or
• 11:30
  2% and your ROI is a hundred percent or actually like an XAI's case, like a thousand percent,
• 11:36
  do you even care? And the answer is of course not. The second thing is you have to remember
• 11:41
  all these hyperscalers were sitting on like more cash together than the U.S. Treasury had.
• 11:45
  So they were sitting on huge amounts of cash. And somehow we're supposed to be sad that they're
• 11:50
  spending it. Right. And they're spending it with like the most absurd ROI we've ever heard of.
• 11:56
  So like somehow some people want you to be bearish about that. Like, oh, my gosh, look at
• 12:00
  they're spending their cash that they had hundreds of billions sitting on cash on hand. And we should
• 12:04
  be sad that they're building stuff with it. And then like, you know what they're doing, right?
• 12:08
  Like you build compute, but then you're leasing it out to customers in like Europe and you're leasing it out to customers from Africa or the Middle East.
• 12:16
  So like our compute is really funding and training the whole rest of the world.
• 12:20
  So we're not talking about like serving 300 million.
• 12:23
  Our compute is serving the whole 7 billion people of the world.
• 12:26
  And so the scale of this is not like domestic.
• 12:29
  The scale of this is international.
• 12:32
  And the ROIs are great.
• 12:35
  And they're not in a lot of cases, they're not borrowing money to do it.
• 12:38
  So they are some of them are like obviously Oracle is you know people like that but like but like a lot of them are using it with their own cash on hand or funding it with their current revenues Right Because like if Microsoft has you know a 50 profit margin that means that they can roll 50 of
• 12:54
  the profits into building data centers. Right. So like it's that I do agree that that spending's
• 13:01
  not going to slow down and the Fed doesn't really have any control over that. And by the way,
• 13:05
  That's a good thing because what we're doing is we're bringing in profits from the whole world to the U.S.
• 13:10
  We're winning the AI race.
• 13:12
  And our shareholders, you know, our people in like my neighborhood in Iowa, like everybody's flourishing from it.
• 13:19
  Right. The stock markets are at rock guard highs.
• 13:21
  And it's something to celebrate.
• 13:23
  And you see like people like Elizabeth Warren, like I actually tweeted, like we need to tax these people and stuff like they're mad that they're doing well.
• 13:30
  And I'm like, how can you be mad about that?
• 13:33
  Like we all want to win is America.
• 13:34
  And that's what we're doing.
• 13:36
  That's what it looks like we're doing.
• 13:37
  And I'm super excited.
• 13:38
  I think it's one of the most exciting times to be investing.
• 13:41
  Yeah.
• 13:41
  And to this point, too, what are the barriers to entry to this this compute leasing market?
• 13:47
  Is it the constraints of the energy grid and access to to power and power infrastructure?
• 13:54
  Like how how how long these ROIs persist?
• 13:58
  Is it because of the supply?
• 13:59
  The infrastructure is still constrained.
• 14:01
  well we have a lot of natural gas in the u.s and so like in iowa what they do we have natural gas
• 14:06
  pipelines that run up from like the shale drillers and what in iowa what they do is they build their
• 14:11
  own generators onto the data centers and then they try to get like wind energy too and things like
• 14:17
  that but like uh generally they're building they're bringing their own power so is it a concern if we
• 14:23
  use more natural gas when we have plenty of it and we're exporting it to the rest of the world
• 14:27
  like i don't think so because it's pretty stupid to put natural gas on a natural gas tanker and
• 14:33
  send it to europe to power a data center over there and raise the cost of natural gas like 20 bucks
• 14:37
  you know what i mean like how dumb is that like it's so expensive to ship it why don't we just do
• 14:41
  the compute here save the 20 dollars you know a cubic whatever of natural gas like it's everything's
• 14:47
  more efficient about it and uh and do the compute closest to where the energy is you know so i think
• 14:53
  there's two things. You need land and you need, but the craziest thing is you've got places like
• 14:57
  California that are just full of empty deserts. Like they won't let them build data centers there.
• 15:02
  I don't know. So we're building them in Iowa, I guess. So Iowa farmland is a little more
• 15:06
  productive than California desert, unfortunately, but that's the way things are. So, but yeah, I mean,
• 15:14
  it's just, it's good for everybody involved and it's, it's going to be really good. I think for
• 15:17
  shareholders, when we start to see the ROI roll in from the data, I mean, we've hardly seen any
• 15:23
  these data centers come online elon musk was like one of the first ones to do it because he could
• 15:27
  build them so fast but like we're going to see in the next year these come online and the revenues
• 15:32
  come in and that's going to be really exciting for shareholders yeah that memphis build out and i was
• 15:37
  telling you before we hit record i was uh been in the bitcoin mining industry for for almost 10 years
• 15:42
  now and watching how they did colossus one by daisy chaining to gen sets and putting the the battery
• 15:48
  powers, the battery walls inside, he was acting like a Bitcoin miner.
• 15:54
  Because a Bitcoin mining, to your point, the whole name of the game is basically shortening
• 16:00
  the distance between the power production and the compute.
• 16:04
  And it seems like AI is adopting that quickly as well.
• 16:09
  And I mean, to that point, it looks like SpaceX is going to build another large data center
• 16:15
  glosses too. It looks like it may be a $17 billion outlay. Yeah. For two gigawatts. So like
• 16:22
  if you can lease out 300 megawatts for 45 billion, how much can you lease out eight times more
• 16:28
  compute for? Like if they only spend 17 billion, it sounds like at the current rate, they could
• 16:34
  lease that out for over a hundred billion. I'm sorry. They could lease that out for over, I was,
• 16:40
  I was way low on that, but they could lease that out for over 300 billion.
• 16:46
  Like, I mean, if you do a 10 X ROI on it, like they're getting on whatever, I mean, it's, it's a lot.
• 16:53
  And how's that going to be for SpaceX shareholders?
• 16:56
  So SpaceX is going to go public in like two weeks.
• 16:59
  You know, they obviously can grow their Starlink division.
• 17:02
  I mean, it provides so much value.
• 17:04
  I think Starlink could easily 10 X their revenues over the next five years.
• 17:09
  And then what can they bring in from compute?
• 17:11
  And, you know, this is really smart.
• 17:13
  They have to build the compute leasing business before they take it to space.
• 17:17
  Because like, you don't want to go to space with no customers.
• 17:19
  So like you build the customer base and then you take it to space.
• 17:23
  And it's so strategic.
• 17:25
  It's so smart.
• 17:27
  And the biggest question I have, like, I think I'm going to be pretty bullish on SpaceX when it IPOs in two weeks.
• 17:33
  But the other question is, what are the best plays to make money off of this, these rising compute like leasing profits?
• 17:42
  And is it is it like the hyperscalers like Oracle and Google and Microsoft?
• 17:48
  Or is it some of the smaller ones like CoreWeave, IREN and some of the Bitcoin miners that are rotating to the leasing out their compute like like Tara Wolf?
• 17:57
  You know, the tickers W-U-L-F and Cypher.
• 18:02
  So I think that's pretty exciting.
• 18:04
  And it's hard to know for sure who are the winners.
• 18:07
  Can the bigger people like Elon Musk just do it faster and better?
• 18:10
  Or can people that are in like Texas and they have really friendly regulations down there
• 18:15
  and they own a lot of the land, can they scale a lot like the fastest?
• 18:18
  So I think a lot of this is going to be who can scale the fastest and get it done quickly
• 18:22
  and have that edge to bring in the revenue first.
• 18:25
  And I don't know what the future holds, but I do think it's going to be good for shareholders.
• 18:32
  Well, I mean, that begs another question, too.
• 18:34
  Like, how many winners could there be?
• 18:35
  I could see there being a lot of winners in this.
• 18:37
  Yeah.
• 18:37
  Because of the insatiable demand for tokens.
• 18:40
  And it doesn't seem like to your point of, I forget which bank was saying it, but the
• 18:45
  gross in token usage just in the last year.
• 18:48
  And if you think about robotics and things that haven't even come to market at scale
• 18:53
  yet that are going to demand probably exponential amounts of compute compared to the nation agentic
• 18:59
  economy that exists now you can see a ton of winners and not only and i think that begs another
• 19:04
  question i mentioned i sit on the board of uh cathedra uh we're in the process of merging with
• 19:09
  a company called sphere 3d but we operate on a lower scale like uh 20 to 25 to 50 megawatt
• 19:15
  I hope you have stock-based compensation.
• 19:19
  Excuse me?
• 19:20
  I said, I hope you have stock-based compensation.
• 19:23
  Well, that's, I mean.
• 19:24
  Some stock options in case it spikes.
• 19:27
  Going back to, I mean, how many winners could there be?
• 19:32
  Could you see a scenario where there's a ton of winners?
• 19:34
  Oh, of course.
• 19:36
  I think that's where we already are.
• 19:38
  So we've already seen a ton of winners in the semiconductors because first they have to order the stuff, right?
• 19:43
  So we've seen all this, like we've seen SanDisk and Micron.
• 19:47
  I mean, Micron was up like 20% the other day, like so many back-to-back 5%, 10% days.
• 19:52
  It's crazy.
• 19:53
  And I think the thing is when you have like a 200% ROI and you really want this stuff
• 19:59
  because you want to get this data center built right now, how much pricing power does it
• 20:02
  give the semiconductors?
• 20:03
  And I think they're figuring that out.
• 20:05
  AMD, you know, NVIDIA, they can continue to raise their prices because these semiconductors
• 20:11
  that are crucial to the compute
• 20:13
  are generating such a high profit
• 20:15
  for the people that have the data center.
• 20:17
  So I see that that profit continues
• 20:21
  and they're just the first ones to see it.
• 20:22
  So the question is like,
• 20:23
  when we built out like the first ones,
• 20:27
  we saw NVIDIA like skyrocket.
• 20:29
  It went up, it's earning like 25X more
• 20:31
  than a couple of years ago.
• 20:32
  Its stock went up like thousands of percent
• 20:35
  because of course it's earning a lot more.
• 20:37
  And then we saw the memory chips go up
• 20:38
  and those went up thousands of percent.
• 20:41
  like SanDisk and Micron. And now we've seen the CPU rally. So we've seen AMD and Intel really take
• 20:48
  off. And so the question is, these are all people that are getting the orders to build the data
• 20:54
  centers and the products that they need to do them. But the question is, when the data centers come
• 20:58
  online, I think we're going to see the data center winners. And here's the thing, almost all the
• 21:03
  semiconductors are doing well. So I could see a scenario where anybody that's been investing in
• 21:08
  the space gets a really high ROI on their data center. So long as they're able to do things like
• 21:15
  house claws, like Elon Musk's Colossus 1 is. So I think it's going to be great. It's just hard to
• 21:23
  know right now because you haven't really seen it in the earnings. So how much do you want to
• 21:27
  speculate to say, I think this is going to happen versus how much do you want to be grounded in the
• 21:32
  earnings and wait for the results to come in first? But by that time, if it turns out to be
• 21:36
  true. The stock will probably already rally quite a bit. And by the way, you've seen it in Cypher
• 21:41
  and Terrell Wolf. They are up a lot. So there's people in on this trade already that are pushing
• 21:44
  these stocks. I mean, I'm pretty sure they're up like 500% in the last year. So there's other
• 21:49
  people that are, you know, that are moving before us, I think on this one. Yeah. And to that point,
• 21:54
  when do you think we'll know for sure that this is happening? You don't want to wait for those
• 22:02
  earning reports? Is this Q3, Q4 this year, Q1 next year? When do you think all these numbers will hit?
• 22:08
  Well, I think like a month ago, people in the industry started to talk about it. And one of
• 22:13
  the people that posted it was like David Sachs. And he's like, hey guys, like you can build a
• 22:18
  $50 billion data center and lease it out for $30 billion a year. And then like, we just learned
• 22:23
  with what Elon Musk did that his numbers were way low, like way low. You can build a $50 billion
• 22:28
  data center and maybe lease it out for $150 billion a year. So he was off by a magnitude of
• 22:35
  five. And that's what's happened, what we've seen in memory chip stocks. And the analysts are just
• 22:41
  like, oh, they'll earn a little bit more next quarter. And then they go and they grow their
• 22:44
  revenue 100% quarter over quarter and just blow away everybody. And so I think it's hard. I think
• 22:51
  the thing is the market's used to valuing growth. That's like 20%, 30% revenue growth year over year.
• 22:57
  nobody knows how to value things that could grow by 25x earnings like and nobody nobody is that
• 23:03
  bullish nobody's like cypher is going to be earning 25x more next year uh oracle is going
• 23:08
  to be earning 30x more next year so nobody really knows how to do that and so i think the market
• 23:12
  frequently especially in tech especially when people can make at low marginal cost so especially
• 23:18
  when you can make additional sales at 100 profit margins nobody really knows how to value that and
• 23:23
  when you think about the history of companies that have done that like google and microsoft and
• 23:26
  Amazon and Visa, they've all crushed the S&P 500 by enormous amounts. And so now we have this next
• 23:35
  stage of these companies that can sell this compute at very high profit margins is what it seems.
• 23:40
  And yeah, do you want to be a first mover? Do you want to wait six months to see it roll into
• 23:45
  the earnings and to be a little less speculative? So I think it's probably, for me, it probably makes
• 23:50
  some sense to do a little bit of both. Go in, but be a little cautious and be ready to double
• 23:56
  down, you know, when the earnings come in. What, what would slow this down? I mean,
• 24:01
  we've seen headlines in the last week, Uber, um, has run through their, their compute budget
• 24:06
  for the year. They ran through it, I believe mid April, you have many other companies saying, Hey,
• 24:11
  the, um, the use of these models isn't helping us add, add bottom line revenue, uh, to, to the
• 24:20
  business is what if, like, what could slow this down? Is it the inability for companies to actually
• 24:25
  integrate these models in a way that makes them more productive, more efficient, most importantly,
• 24:30
  more profitable? What could pop a hole in this, this gross story right now?
• 24:37
  So I think the interesting thing is anybody buying these tokens in the first place has probably
• 24:41
  already really done the math on it because you'd be pretty stupid to just buy tons of them and make
• 24:45
  it more expensive than human labor. Okay. So like, imagine this, you're sitting in a data call center
• 24:50
  And your job is literally just to answer the phone and take customer service calls.
• 24:57
  So they're paying everybody $25 an hour.
• 25:00
  You have thousands of employees sitting in this call center.
• 25:03
  You could have a chatbot that's like, oh, okay, we've seen that you got a fee on this
• 25:10
  credit card or this payment was fraudulent.
• 25:13
  We're going to go ahead and reverse it and send you a new card.
• 25:16
  You don't actually need a human to do most of this, probably 90% of this stuff.
• 25:20
  So almost every customer service call you take today that basically done by a token That like a tokenized chat box So like with your credit cards with your banks pretty much all that they buying tokens to
• 25:34
  fund what they're doing, to fund these agentic AI agents that are carrying out tasks that humans
• 25:39
  used to do. So like, first of all, I don't think that they're doing that because the ROI is bad.
• 25:46
  And then the second thing is in software development and like Uber and Salesforce,
• 25:50
  they're buying huge amounts of tokens too, because they can write 10 times the amount of code for one
• 25:56
  tenth the cost. And so they're doing it because the ROI is really good on it. And they're the
• 26:02
  first mover. So you have to remember when like somebody in New York city or Silicon Valley is
• 26:05
  doing this, the rest of the country and the rest of the world isn't even close to being where they
• 26:09
  are and adopting it. So like you see like the first 1% moving, you don't see the other 99% yet.
• 26:16
  And that's why the token demand is like projected to just explode. Like when I'm sitting here in
• 26:20
  Iowa. I don't know anybody that uses them. But I talked to my friends that I went to Northwestern
• 26:26
  and University of Oxford with, and they work for some of the tech companies and stuff. And they're
• 26:31
  all like, yeah, we use this all the time every day. And so I think we've seen the early adopters
• 26:36
  move, but we haven't seen the rest of the 99% of the world. And the rest of the world and the rest
• 26:42
  of the people in Iowa and everything, they're going to be buying those tokens from these companies.
• 26:47
  and the tokens are going to be made in the United States.
• 26:49
  They're not going to be made in Europe
• 26:51
  where natural gas is 50 times higher.
• 26:53
  They're not going to be made in Africa.
• 26:54
  They're not going to be made in China.
• 26:56
  So like, we really are the answer to the whole world here.
• 27:00
  And you cannot, like, do you remember like Google,
• 27:03
  like $100 billion market cap?
• 27:05
  And people are like, oh, it's so big.
• 27:07
  It can't grow anymore.
• 27:08
  You know what I mean?
• 27:09
  And they just forget that Google can go
• 27:11
  and sell their AdWords,
• 27:12
  the entire rest of the world, 100% profit margins.
• 27:15
  Like nobody can even fathom that the demand will actually be 50 times higher at 100% profit margin.
• 27:21
  Same thing with Microsoft.
• 27:22
  Like, and so they just underestimate it.
• 27:24
  And then the stock goes up, what?
• 27:27
  10,000%?
• 27:28
  Like, so I think we're probably in the same scenario here.
• 27:33
  I just think it's like, you know, I think it's just the biggest threat would probably be that Anthropic tries to build their own data centers.
• 27:42
  But I still think that they're so far behind the ball and the demand is so big that they can't possibly do that.
• 27:48
  And they don't even have the capital to do it.
• 27:50
  So I'm not sure.
• 27:51
  I'm not sure what could deflate this.
• 27:54
  But right now, I'm not thinking not much.
• 27:57
  I'm thinking that there's not much that could deflate this at the moment.
• 28:00
  I mean, there's political opposition against it.
• 28:03
  But isn't that even better?
• 28:05
  Like if the rest of the world doesn't want to build them because there's political opposition and the red states will.
• 28:10
  you know like isn't that better that we have like the place where we can break the red type and put
• 28:15
  them up faster than anybody else anywhere in the world so like that that just gives us a bigger
• 28:19
  moat and it makes our it makes investing in them safer yeah i mean two points to add there i think
• 28:25
  when it comes to implementation of these ai workflows or agentic
• 28:30
  workflows it really comes down to being able to implement i think jack dorsey
• 28:36
  in Blocker, a great example. I've watched their implementation of AI into their company over the
• 28:43
  last year, and I think they're doing it the right way. And I think Jack's interview with the Sequoia
• 28:47
  Partners a couple of months ago really highlighted how they're viewing it. And I think the approach
• 28:51
  to how you actually implement these AI systems is very important. There's going to be a ton of
• 28:56
  companies who don't think strategically about those implementation details that mess it up,
• 29:03
  and they wind up wasting a bunch of money.
• 29:05
  And then on the other hand, like I think for small teams,
• 29:07
  like my humble media company that we're running here,
• 29:10
  we've been running an open claw agent since January.
• 29:13
  No way.
• 29:14
  It has expanded our ability to do what we do here
• 29:20
  by orders of magnitude.
• 29:21
  And I think for small teams,
• 29:23
  it's going to be particularly powerful.
• 29:26
  And then when you think about the profit margins
• 29:27
  that can be gained by small teams that implement this from zero,
• 29:31
  the profitability of the smaller team is going to way exceed the larger companies that have a ton
• 29:38
  of bloat. And I think that's going to be actually an interesting theme to watch moving forward is
• 29:44
  the incumbent behemoths that were using headcount as a KPI back in the easy money days of the 2010s
• 29:51
  and early 2020s, trying to unwind the mistake of that KPI. Yeah. I mean, how amazing is it that
• 30:00
  like Elon Musk went in and like cut 80% to X and it worked just the same.
• 30:05
  You know what I mean?
• 30:06
  And I do think there's a thing of like,
• 30:07
  when people throw crazy money out at things,
• 30:11
  there's like kind of infamous for getting ripped off.
• 30:14
  And so what companies are run so well that they can build them the leanest.
• 30:17
  And I think they'll have a huge competitive advantage, right?
• 30:21
  To build them the fastest and the leanest.
• 30:23
  So you probably, honestly,
• 30:25
  it sounds like maybe you know more about some of these like data center
• 30:28
  place and it's amazing that you're in this space on the board of a company so uh so you're gonna
• 30:33
  know better than anybody but it sounds like you don't want to scale back your clawed bot and it
• 30:37
  also sound or your open claw it sounds like you also like so you you know you would maybe use more
• 30:42
  but also i would say you're a first mover because you're like a high-tech guy so like how soon until
• 30:47
  everybody else figures out what you're doing and how great it is yeah and i mean that's the thing
• 30:52
  Right now, the other thing you have to consider, too, is these models are progressing so quickly.
• 30:59
  So you'll build something one week, and then a new model will be released.
• 31:02
  It's like, okay, I have to take this out, put this in, and rejigger things a certain way.
• 31:07
  And it's a lot of learning on the go.
• 31:11
  And so to your point about the early movers, and I'm not trying to blow smoke up my own ass,
• 31:15
  but it is an involved process.
• 31:17
  And if you're not staying up to date with the latest developments in the space, it's going to be hard to keep up.
• 31:23
  But there will be a point, I think, where the model, the return on the model, sort of the model progression sort of plateaus.
• 31:33
  maybe doesn't plateau, but it becomes more stable of a rise in efficiency gains and,
• 31:39
  yeah, I guess the IQ of the models themselves that you'll be able to create out-of-the-box
• 31:47
  solutions.
• 31:48
  Like setting up an open claw in January was extremely hard.
• 31:52
  It's easier today.
• 31:53
  I imagine a year from now, there will be a point and click out-of-the-box sort of agentic
• 31:59
  flows for companies that are very easy to set up. Yeah. I mean, and if demand grows by 24X
• 32:08
  or 80X, we don't have anywhere near amount of compute. Even what we're building today,
• 32:15
  it's nowhere close. So it honestly seems like it could be like an eight or 10 year build out to
• 32:21
  catch up, which means we'd be in the very early, I mean, like in the first like, you know, year or
• 32:26
  two of a decade long build out. So it could be one of the longest CapEx cycles that's ever been
• 32:32
  in the history of the world. I mean, it can really only be compared to when they needed to build
• 32:36
  railroads across the whole world because they realized that was better than taking stuff on a
• 32:40
  donkey. Yeah, well, I think many people are comparing it to the internet and the broadband
• 32:46
  build out too. And to your earlier comments about people worrying about this being a bubble and it
• 32:52
  being too hard to believe and they're pointing at the dot-com bubble i think that's one thing that
• 32:58
  many people are missing is that the demand for tokens is there it's like the the dot-com bubble
• 33:02
  the infrastructure was built out and we didn't have the means to really lean into the digital
• 33:09
  economy that exists today in the 90s and it didn't you weren't able to monetize websites the way you
• 33:15
  are today there's a ton of sort of software development kit infrastructure and payments
• 33:19
  infrastructure and logistics infrastructure that needed to be built out and Amazon and Google and
• 33:25
  many others did that in the late 90s, early 2000s. And then that caught up. But I think with AI,
• 33:30
  it's different because the models, they work. If you know how to use them, they work well for you
• 33:35
  and you want to use them more, but the tokens simply aren't there yet. The ability to ping the
• 33:42
  models as much as you want isn't there. Whereas in the broadband era, the fiber was there. It was
• 33:50
  laid, but the supply of people bringing businesses that actually worked on the internet wasn't there.
• 33:57
  Yeah. I mean, and I think the thing is, it's like, it's kind of like the railroad,
• 34:00
  like Wells Fargo started as a carriage delivery, like a wagon delivery service.
• 34:05
  And so like they'd go down the Oregon trail from whenever in New York, all the way down to Oregon.
• 34:09
  And it's like the railroad basically put that out of business overnight and there became a bank.
• 34:14
  But it's like, if you can do something for one one thousandth the cost, you know, and software has always been limited by extremely expensive developers sitting in seats.
• 34:23
  If you can like lower things by one one thousandth of a cost, do you really ever go back to the way things were?
• 34:30
  Is there really any slowing down the adoption?
• 34:32
  Because its savings are so big, it only accelerates.
• 34:35
  And I think that's probably where we sit today.
• 34:39
  Yeah. Outside of compute infrastructure specifically, again, bringing it back to energy, because that's one thing I've identified as maybe the biggest constraint right now is energy infrastructure.
• 34:54
  And I talked about natural gas pipelines, but I think we really need to expand generation and grid interconnectivity by orders of magnitude in the next decade as well.
• 35:04
  And I think that's another part of the market that provides an incredible opportunity.
• 35:07
  and it's just well the data centers aren't so i think if the data centers are bringing their own
• 35:14
  energy and they barely connect to the grid then one is that true and two and two like there's
• 35:22
  people like bloom energy making like um like obviously they make electric generators right
• 35:27
  so that you can like bring your own power to uh to a data center but like the question to that is
• 35:33
  is that like, that's really expensive to build each data center.
• 35:38
  So, or to build each generator, I mean.
• 35:40
  So the question is, like, that's not low marginal cost, right?
• 35:43
  Like you can make an additional memory chip for not a lot of costs,
• 35:46
  but it's really expensive to build a big generator.
• 35:49
  So the question is, is the earnings growth going to be the highest there?
• 35:54
  And like producing electricity is always going to have a high marginal cost
• 35:58
  because it costs a lot to make, you know,
• 36:00
  whether it's a natural gas or wind turbines or solar,
• 36:03
  it actually costs something and each additional unit you make is not free. It really has a cost
• 36:07
  associated with it. So I don't think the ROI is going to be the highest in that sector for that
• 36:12
  reason. Doesn't mean it won't be good. Doesn't mean it won't be good. I think that you can make
• 36:16
  more money in semiconductors or something, in my opinion. I think the other thing that I think has
• 36:22
  really been overlooked is software in this. And I know the software, these are some of the people
• 36:28
  buying the tokens. And so there's some increased costs there because they maybe necessarily haven't
• 36:34
  scaled back their workforce. So they've kept all the high costs, but they've also spent all the
• 36:38
  money on the tokens. So they were like kind of double slamming themselves. But software has
• 36:45
  always been limited by the number of subscribers sitting in a seat, right? So manpower. Manpower
• 36:50
  is expensive. So here's the thing. You open your open call and it's doing advertising work for you
• 36:57
  and it's working in Adobe, and maybe you have one doing sales outreach for you, and it's working in
• 37:01
  Salesforce. Software is no longer limited by the number of people you have. So now you could have
• 37:08
  one person and they could have 20 subscriptions, right? Like you could have the same thing,
• 37:12
  drawing CAD drawings and Autodesk, right? And so now the question is, does demand get totally
• 37:19
  unleashed? It was always limited by expensive manpower. And now software demand is totally
• 37:24
  unleashed because AI agents need the tools to get the work done. And so I think that could be
• 37:31
  really interesting in that the demand is actually going to be higher. Additional sales come at 100%
• 37:38
  profit margins. And you have 5X demand or something as the tools just become... Because
• 37:44
  the most expensive thing to Adobe was never a subscription cost of 300 bucks a year or 1,000.
• 37:51
  It was the person that was, you were paying to do the work in it.
• 37:54
  You know what I mean?
• 37:55
  They might cost a hundred thousand.
• 37:56
  So if you're lowering the cost to use these by a magnitude of a hundred, doesn't demand
• 38:02
  go up?
• 38:03
  And because that additional sales come at high profit margins, doesn't earnings like explode?
• 38:07
  And so I bullish on software Like I hold it in AOTG You know I topped like some of the top holdings or you know besides semiconductors and besides some of this AI infrastructure are names like AppLovin and Microsoft and Toast
• 38:23
  And so I think it could really be unleashed.
• 38:25
  And I think that the potential there should not be understated.
• 38:29
  You know, and there's downside, of course, that some of them are losers and always keep that in mind when you invest.
• 38:34
  But when everybody focuses on the downside and nobody focuses on the upside and the PE ratios are like 15 and 20, and they're really reasonably valued with very robust earnings growth still today.
• 38:50
  I think that sets up to some of the best trades.
• 38:53
  And you saw that a year ago with memory.
• 38:55
  Everybody was bearish on it.
• 38:57
  And the PE ratios were really low and the growth was good.
• 38:59
  And I think when you have those things and you combine those all together, those are the makings of the best trades.
• 39:09
  And so we'll see what the future holds.
• 39:11
  Anything can happen, but I am excited for it.
• 39:14
  Yeah, I think we saw an example of this a couple of weeks ago when Figma announced their Q1 earnings, and they surprised massively to the upside.
• 39:21
  I think many people were thinking Figma UI designer tool was going to get blown out by people just building it themselves.
• 39:30
  But they actually implemented some AI functionality.
• 39:33
  And I think to your point, what they experienced is people actually using them more because they have these tools to go in there and leverage their software tools.
• 39:43
  Yeah, well, here's the most interesting thing is a lot of these companies have like a 5-10% profit margin.
• 39:48
  So imagine you're a company like Spotify, and you can look at the earnings today,
• 39:52
  and their year-over-year operating expenses are falling.
• 39:55
  Meanwhile, they're still growing their revenues at 10%.
• 39:57
  So what that tells me is that their profit margin will go from 5% to all the additional
• 40:04
  10% sales go straight to profit.
• 40:07
  So now their profit margin will be 15%.
• 40:09
  Well, that's 200% year-over-year gap earnings per share growth for a company that had a low
• 40:13
  PE ratio.
• 40:14
  So I mean, usually when they have low PE ratios, they shouldn't get any lower.
• 40:17
  You know, like usually they should keep up with earnings growth.
• 40:21
  And so like making 200% in a year is a big deal, like in any stock, right?
• 40:25
  You know that.
• 40:26
  And so how many companies today like AppLovin and Robinhood are implementing?
• 40:32
  I mean, Robinhood's financial services, but what are they really?
• 40:34
  Like they're really proprietary software to implement brokerage services.
• 40:38
  So like how many of these companies are going to keep their expenses constant and grow their
• 40:42
  revenues, not just 10%, but maybe 50%.
• 40:44
  50%. And like AppLovin is obviously growing at like 60%. And that's really, you know,
• 40:51
  that really the upside of that potential, I don't think can be overstated.
• 40:56
  Yeah. So you're not a doomer at this. You don't think AI leads to massive
• 41:00
  job disruption that completely throws the economy into a tizzy and destroys the demand for tokens
• 41:07
  that existed in the first place? Do you think this is going to be an accelerant for human
• 41:12
  productivity and creativity. Yeah. Yeah. I mean, it brings in a lot of wealth. Like I just look at
• 41:17
  what it's doing in Iowa and every, it seems like people are doing better than ever. And, uh, so
• 41:22
  I mean, I think in the 1900, they were building the Panama canal and they built, they came out
• 41:28
  with a steam shovel and there was actually a political debate at the time with like the
• 41:32
  presidents and stuff. Should we bring this steam shovel to Panama canal because a bunch of people
• 41:37
  lose their jobs or should we just keep paying all these people to die of malaria in Panama?
• 41:41
  and keep doing it because we want them to be employed.
• 41:45
  And is that not crazy that they were considering
• 41:48
  digging the Panama City Canal with a shovel
• 41:50
  because they didn't want to fire people?
• 41:52
  They didn't want to reduce the workforce
• 41:53
  and have a clause like a political firestorm.
• 41:55
  And then they did do the steam shovel.
• 41:57
  And guess what?
• 41:57
  Everybody went and got better jobs
• 41:59
  and provided more value to the economy.
• 42:01
  And so I think that's where we are today.
• 42:04
  Somebody that's just doing payroll right now or something,
• 42:06
  or they're doing accounts payable,
• 42:08
  and it can really be automated.
• 42:09
  they can go get a better job doing something else and they'll add more to the GDP than being able to
• 42:15
  automate that task. So like when you add more to the GDP, you increase the output, like the country
• 42:20
  gets wealthier and as a whole, everybody gets wealthier together. So I'm really bullish on it.
• 42:25
  I'm really bullish because nobody's just going to go and sit at their house. They're going to go and
• 42:28
  do something like that. Like you and I would go make a company. You know what I mean? Like, so
• 42:32
  like, like people go do that. They're not just going to sit around in general, I think the
• 42:36
  majority of people and then they'll provide value and it'll, you'll see a GDP explosion,
• 42:40
  I think. And you'll see some of the best growth maybe that's ever been seen.
• 42:45
  Yeah. Is there anything, um, on this AI topic, uh, that we haven't covered yet? That's top of
• 42:51
  mind that you think the audience should be aware of. We've been, we've been whacking all the moles
• 42:56
  on this one. So I'm trying to think, but you got me in the corner now and trying to make sure I
• 43:00
  don't leave anything out. Um, yeah. Um, you can't think of anything, man.
• 43:08
  Well, on that note, moving to fertilizer, since you, since you're close to the, uh,
• 43:14
  the fertilizer, fertilizer industry as well. I think there's been a lot of headlines about
• 43:18
  obviously the straighter from use, uh, messing up the supply chain for, for sulfur and helium,
• 43:24
  which are key inputs to, to fertilizer. What are you seeing on, on that end?
• 43:28
  Yeah, it's not great. So now we're taking a little worse of a step. But I think there's three things to keep in mind. I won't hit on helium. So you've got nitrogen. The way you make nitrogen is you use a ton of natural gas to superheat the air and cause a catalyst and you take the nitrogen out of the air.
• 43:49
  So it's like naturally occurring, but you take nitrogen out of the air, the atmosphere is 70% nitrogen, and then you turn it into like ammonia or UAN.
• 43:58
  And that's what you use to put on corn and things like that.
• 44:02
  So 30% of the natural gas flows out of Qatar and it goes to places like India and China and Europe.
• 44:09
  And so they can't get their natural gas.
• 44:11
  So all of their nitrogen plants shut down.
• 44:14
  So they're not like making very much nitrogen over there.
• 44:17
  uh so that's a big problem because they made like the majority of the world's supply
• 44:22
  so the u.s actually is a net importer we have a lot of plants but we actually bring in more than
• 44:26
  we export so now the u.s producers are sitting there making very like they have uh you know
• 44:32
  they're obviously using buying natural gas at like three bucks and they're making it
• 44:36
  and everybody else is like you know getting slammed on it in the rest of the world so
• 44:40
  the u.s people like cf industries and uan stock they're all doing really really well like great
• 44:45
  years, everything, but it, there, but it's because there's such a tightness and they can export it to
• 44:50
  the rest of the world that raises the prices here in the U S that's bad for the farmer because they
• 44:54
  have higher input costs. So I think that like, that's, that's number one, nitrogen prices are
• 44:59
  up like a hundred percent. So that's tough. Uh, that's tough. And corn's not up a hundred percent.
• 45:03
  So that's tough for the farmer. The second thing is that sulfur. So sulfur, you actually put on
• 45:09
  corn, like in that price is going up, of course. So, uh, so that's bad. But the second thing is
• 45:14
  you like try catalyze sulfur into sulfuric acid, and then you use sulfuric acid to melt phosphate
• 45:20
  ore. So when you're mining phosphate, you melt the ore and that makes a phosphoric acid. Well,
• 45:26
  phosphoric acid is the base of like phosphate fertilizers. And so now the sulfur prices are so
• 45:34
  high that they can't make phosphate profitably. So they're shutting down. Like Mosaic is a public
• 45:40
  glycic company, they're shutting down production in their phosphate plants because they can't get
• 45:44
  enough sulfuric acid. And so phosphate prices are now going through the roof and sulfur prices,
• 45:50
  which is another key ingredient. And so you've got sulfur, phosphate, and nitrogen that are all
• 45:54
  spiking really high in prices. At the same time, you don't see corn really necessarily reflecting
• 46:00
  that rally yet. So it's creating really tough farm economics. And what's really interesting for us
• 46:05
  is that our customers, so what we do at Twin State Inc., you know, has started my grandfather,
• 46:11
  is we buy nitrogen, we buy phosphate, and we buy potash and sulfur and zinc from people that make
• 46:18
  it. And we turn it into, like when you fertilize your yard, you get a blended fertilizer, right?
• 46:23
  So you get something that has all five of those ingredients and more, and then you put it on your
• 46:27
  yard. Well, we're the people that blend it all together. So we have to pay the higher in-book
• 46:31
  costs also. It's not like we're like, oh my gosh, prices are higher. We're making more. It's exactly
• 46:36
  the opposite. Prices are higher and our quantity being sold is dropping because people are having
• 46:41
  to cut back because the prices are so expensive. So I mean, it's tough there. It's hard on the
• 46:46
  farmer. What we really need to see is a rally in corn. And then the economics will make sense again.
• 46:52
  And I think that's going to happen. It just takes time to have the commodities market react. These
• 46:59
  things are cyclical, but sometimes it takes a year to see reactions as you draw down stockpiles,
• 47:03
  things like that. So we'll see what happens. But not the best year for ag, I'd say.
• 47:10
  Have you experienced an environment like this before as your family?
• 47:16
  Not me. The farm crisis in the 80s was bad, and we're not near that. But that's when a lot of
• 47:22
  people are going bankrupt. But I would say that usually all of our customers use John Deere
• 47:26
  financing, which is, you know, kind of thought of as being very easy to get, uh, because John
• 47:31
  Deere will like lend, uh, you know, they'll, they'll lend and they'll finance. They actually
• 47:34
  make more money financing, I think from John Deere than they do from their tractors. Nobody
• 47:37
  seems to know that. But, uh, um, but, uh, this year John Deere is saying like, oh, you guys are
• 47:44
  so stretched. We won't lend to you. So they're going to like non-traditional lenders and much,
• 47:48
  much higher interest rates. So that's like, I mean, we might be seeing 40% of our, our customers
• 47:55
  that we sell to doing that. So that's not good to see. That means that people are distressed
• 47:59
  and the assets are so... The liabilities are becoming more than the assets. So traditional
• 48:04
  lenders are no longer lending to them. And that's usually what you see before you see bankruptcies,
• 48:10
  because after that, then obviously they can't pay the interest rate and they go bankrupt.
• 48:14
  So I hope we don't get there. I hope we see corn go up and the incomes go up and then everybody
• 48:20
  gets out of a tough spot. But you could see makings of a crisis here that could be similar.
• 48:27
  Yeah. Very interesting times. We got the incredible optimism and growth potential with AI and then
• 48:34
  some supply chain disruptions, messing up the food supply. But I think we'll learn anything
• 48:42
  from 2020, 2021 specifically is the supply chain disruptions can be fixed in time.
• 48:50
  Yeah. And part of the problem is that land prices just had gotten so high, right? So people kept
• 48:56
  buying higher and higher land and you need a higher profit just to cover the cost of the land.
• 49:00
  And then they rent it out for more. And so some of it is just that the asset prices had been so
• 49:05
  high, kind of like how you could see like, you know, home prices getting too high in an area.
• 49:10
  And then they kind of stall out for a while and fall.
• 49:12
  And, you know, I think there's some of that going on, too, which is that this was like farming had been so safe.
• 49:19
  The land had been such a good investment that it was so reliable that just more momentum money just kept pouring into it and overpriced the land.
• 49:26
  And that exacerbated, you know, a tough time because rents are high and, you know, interest that you're paying on overpriced land is high and all that stuff.
• 49:36
  Yeah.
• 49:36
  Yeah. Maybe the agents can help us figure out a solution to these problems.
• 49:42
  That's good. I hadn't thought about that one yet.
• 49:45
  Awesome. Well, John, again, thank you for hopping on a short notice. This was a fascinating, highly dense conversation.
• 49:54
  And I have a lot of people on here who are very, very skeptical of the AI thing, but just using it myself over the last couple of years, but with the agents over the last five months, it's become pretty obvious to me that there's a there there.
• 50:10
  It's real.
• 50:10
  If you know how to use it, it can make you a better operator.
• 50:14
  And it's just going to be there's just going to be a lag period before the broader market wakes up to that begins implementing this.
• 50:24
  And if that happens, the demand for these tokens is going exponential.
• 50:28
  We're at the cusp of what we're going to see in the next decade.
• 50:32
  Yeah, well, you've got a great show.
• 50:34
  You provide enormous value to your viewers through your insights.
• 50:37
  So I'm just happy to be a part of it.
• 50:39
  And thanks for having me on.
• 50:40
  All right.
• 50:41
  We'll link to John's X account, what he's doing at AOG in the show notes.
• 50:47
  And hopefully we can do this again soon.
• 50:49
  Yeah, yeah.
• 50:50
  So thank you.
• 50:51
  And yeah, check out AOTG ETF if you guys are interested.
• 50:54
  Awesome. Peace and love, freaks.