The AI Bubble: Boom, Bust & Baked in Nimbin

0:00You want to get your backyard summer ready, but you don't want to break the bank? Wayfair gets it. Planning on dining al fresco or relaxing poolside? Wayfair has everything you need to prep your space. Shop now and save up to 70 % off during Wayfair's 4th of July clearance. Score huge deals on outdoor furniture, area rugs, and more. We're talking thousands of products for every style and budget. Plus, surprise flash deals July 6th. Don't wait. Shop Wayfair's 4th of July clearance now through July 6th at Wayfair.com. Wayfair, every style, every home.

0:33David McWilliams: to understand the economy you have to understand human nature this podcast is powered by a cast it's time for the podcast the australian odyssey continues john is looking at me down the tube i have a head that looks like a nuclear power station it's glowing amount of energy i've had to put you know those screens on my screen just to dim it down a bit i john yeah yeah yeah you know the way they always say look even if it's overcast in australia we're factor 50 i know i know i know so now i i'm just showing john down the line by two bottles of aloe vera as i try and sustain you'll be through those by the end of the podcast i know it was dreadful and Well, I did a gig last night and I was just like this big sort of tomato head on the stage.

1:31David McWilliams: The lights on you. Because you're doing, yeah, the lights are on you. So you're just like reflective. Anyway, it's great to have you all. As you know, the journey down here continues. But today's podcast, we're going to talk about the AI boom, the potential AI bust. What is going on in AI? What are the financials? But interestingly, we're going to make the point that the people behind AI, the Silicon Valley mind, actually wants there to be a bubble. So there are three types of ways of looking at AI. First is we know that in the first half of this year, the super scalars, what they're called, the companies behind AI.

2:18David McWilliams: So you've got everything from NVIDIA that makes the chips to Meta and Google that need these data centers to create the data from which AI can scrape. Together, it is estimated that these companies are, and of course, OpenAI, the Sam Altman company, are going to spend between three to four hundred billion dollars. Billion dollars, right? Yeah. Yeah. It's going to be half the growth rate of the United States this year is going to be wrapped up in AI, which also gives us the warning sign that if AI collapses, what happens to the growth rate of the United States, particularly given the election in 2028, because we are in what I would call a sort of a Kindleberger and Minsky boom-bust cycle.

3:11David McWilliams: It's so clear to me, right? Now, what is fascinating, John, is this is also clear to the AI boosters. All of them, even if you heard Jeff Bezos last week said, this is a bubble, but it's an industrial bubble. It's a good thing. Yeah, I'm not sure what that means. So what I want to make the point is that even the people involved in AI want this to be a bubble. So what I'm going to say is that, yes, there is a bubble. yes that bubble is going to burst at some stage but what the AI boosters are saying is that AI is a transformative technology a once-in-a-generation technology not unlike the railroads are the telecom boom of 20 years ago so yes people will lose money but what you need is this irrational exuberance to propel forward people.

4:03David McWilliams: And I got a book last year called Boom, Bubbles and the End of Stagnation, which was written by a guy called Byrne Hobart, an American guy. But it was published, interestingly, by Stripe Press, the press and publishing arm of Stripe, the Collison Brothers company. And what this suggested was that the techno guys want bubbles, because it's only through bubbles can you break out of the straitjacket of stagnation. So this is a fascinating thing, which means that we in the financial world or the economics world are worried about the financial fallout of a bubble. Democrats should be worried about the social fallout of a bubble because what we know is that contagion, financial contagion, tends to be like almost like a Death Star that pulls lots and lots of people into it when things go pear-shaped.

4:59David McWilliams: So it'll have huge ramifications, even for the Trump presidency, if this thing actually goes pear-shaped, right? Yeah, absolutely. But the Silicon Valley guys, the engineers, the technologists, want the bubble because they celebrate the bubble, because for them, it's only money, after all, which can be printed. Yeah. Seriously, and it's actually other people's money, in many cases. We're going to talk about this. Well, that's more to the point, yeah. Because they look at things like the Manhattan Project, the Apollo project, and what they see is these moments of great American acceleration, where, in fact, accountancy and worrying about finance and things are the least of their worries, because what they basically say is the Americans would have never gone to the moon had they been worried about if they had been encounters saying, oh, look, don't spend that money on this thing, right?

5:45David McWilliams: So we have three different types of mindsets here. We have the technologists who want the bubble, even though they understand and they appreciate that certain people are going to lose. It could even be them. We have the finance people who are terrified of the bubble because they have leveraged themselves to this dream of an AI future. But of course, we've got the democratic process in the middle that is observing all this almost from the sidelines. But we know that when bubbles burst, the financial, social, political and economic fallout is never limited to the sector itself. Why? Because finance and money is contagious.

6:30So when you get a risk environment where nobody's worried about risk, everything can get financed.

6:38David McWilliams: But if there's an end to a bubble, i.e. the AI bubble, what will happen in the risk profile of the whole society changes, which is exactly what happened in Ireland when the housing market crashed. Banks didn't lend for anything because the banks went bust themselves. So this is what we're going to talk about. And the added thing, John, is the fact that risks are being increasingly concentrated on both the spending and the financing side. And that's very important to explain. So on the spending side, what you have is very few players, access to vast amounts of capital, spending in one specific area.

7:16Yeah. Data centers, chips, data centers, et cetera.

7:21David McWilliams: But on the other side, you have an increasing concentration of financial providers, which is the new kid on the block, private equity. and private equity and private credit is getting into bed with the major, what they call super boosters, the metas of this world. They are therefore increasingly concentrating their capital risk in this area. Meta is increasingly concentrating its capex risks in this area. Everything is now opaque. They're beginning to produce all sorts of weird sort of special purpose vehicles for this business. But yet, John. Yeah. the investment will benefit humanity, like the railroads and all those things.

8:07David McWilliams: What you will find is that, you know, previous expansions, even if people lost money, they still create technology that benefits everyone. Well, do you know the interesting thing about it is, and when I talk to people about technology in general, is that, you know, it's been foisted upon us. Now, I obviously use technology in my job all the time, and the stuff I just can't do without. But then there's the other kind of non-worky technology that I love. And here's a really good example of this. Okay. The other night, I got up in the middle of the night because I'm of a certain age, and I just checked my phone.

8:46And at four o 'clock in the morning, my time, one o 'clock in the day, lunchtime in Australia, and I look at my phone and there's a WhatsApp text on it. Let me read it to you.

8:58David McWilliams: By the way, this is on a WhatsApp group of our mates. Now, our mates are of a certain generation. A, we get up in the middle of the night to go to the loo. But B, because we were born in Ireland in the late 1960s, we are scattered all over the world, right? That's right. That's right. So on our WhatsApp group are friends of ours who live in New York, friends of ours who live in Australia, friends of ours who live in Hong Kong, and of course, a whole load of people who live back in Dublin and loads of people who live in London. So that's a... And everywhere in between. And so I see this message from your man over there, and it reads, these little bastards are stronger than you think.

9:32Greetings from Nimbin, is it? Nimbin? Australia. Weed Central. Google it.

9:39David McWilliams: Okay, so let me introduce. So Nimbin is a tiny little village, a hippie village in Australia, in the north of New South Wales, beside a place called Byron Bay. And because it's a hippie village, it is the weed capital of the world, or certainly in Australia. So, of course, I decide to go there to have a look. And I walk into this place, and I'm not a weed smoker, John, as you know, but when in Rome, do it the Romans. Absolutely, absolutely. So I decide to buy a little gummy just to get me going. So I eat the head off the gummy. and then about 20 minutes later this is what I sent in a quick WhatsApp oh my god I took a photo of the gummy and I sent it to the lads and say these little fuckers are stronger than you think but this is the message that so Pat comes back when he mentioned Nimbin oh I know well and Matt goes any tips I'm a bit splattered now but we'll be grand in an hour or so that wasn't true that wasn't true Matt sure wasn't so then Peter Kavanagh who's a day behind, actually.

10:50Behind? Yes. He's in New York. Yeah, he's midnight in New York. And he's going, Jesus, be careful, Matt. Calm down a bit. Calm down. So Pat comes back with some, you know, sage advice on what to do as a stoner in Nimbin. And Matt comes back with, oh, my, head down there later. At the moment, it's hard to go to the bar, let alone contemplate leaving Nimbin for the coast. too stoned to even get up out of my seat. So I'm looking at this and I'm going, like I'm going back to bed. And I said, fancy doing a pod? Peter Kavanaugh comes back really urgently going, no, no, no, man, that's a really bad idea.

11:32I think we should.

11:33David McWilliams: And the funniest thing is, it's a good idea. So basically what happened, right? This is how technology changes the world. Yes, exactly. This is the technology I like. So, dear listener, I'm sitting there a little bit. By the way, I was really grand sitting on my own, but then I took a bite out of the chocolate-flavored mushroom cake and then the wheels fucking came completely off. Oh, my God. Let's just say it was a bizarre day. I'll tell you all about it. Anyway, my point is, right, I got this great idea for a podcast when I was stoned there. I thought, look, this is it, right? You have an intercontinental WhatsApp group, which is unifying old friends at various different times of a 24-hour clock around the world.

12:24David McWilliams: We are exchanging notes, warning each other, a certain pastoral care towards each other. Yes, there is. There was a lot of empathy there. A lot of empathy, right? But there's also little information that I'm giving that if you're in this part of the world, you might come up to this Nimbim place just to have a look at the whole idea. Now, the point was, WhatsApp is the product of the last bubble. WhatsApp is an integrated telecom system which wouldn't exist had people not spent a fortune in the last 20 years on telecoms, on all sorts of technology. So even though people lost their shirt on the telecom boom 20 years ago, the legacy of the telecom boom is the fact that we have intercontinental instantaneous communication.

13:11David McWilliams: And that is broadly a good thing. So as I was baked in Nimbin and you said, let's do a pod, I thought, actually, this precipitated an entire train of thought, which leads to today's podcast, which is AI. Because in the same way as WhatsApp allowed me to share some essential information to you gentlemen, over the course of the 24 hours that I was there, which really actually blended into about two minutes. I can't remember a thing. But anyway, there you go. The key idea is that technology leaves positive legacies. Absolutely. Those positive legacies are emblematic in our WhatsApp group. Now, many listeners might say that they're not particularly positive.

13:57David McWilliams: These are not life-changing. Well, that afternoon they were life-changing, life-affirming, right? And more importantly, to see the present AI boom through the context of the long-term benefits vis-a-vis the short-term bubble. And that's as I was sitting in a bizarre cafe in Nimbin, looking at the tie-dye hippies and thinking to myself, God, I wouldn't like to be one of them. And then realizing, fuck, I am one of them. and getting involved in conversation. I got involved in conversation with a biker group. Oh, really? Yeah, yeah, yeah, yeah. How did that work out? For some reason, they took to me.

14:41David McWilliams: So you've like, you know, the economist and the biker walk into a bar. You were their little toy, were you? I was like, exactly. But when you're talking about tech, though, I mean, you know, prior to this episode, you know, we've been able to do podcasts while you were in Australia in all sorts of, you know, and I'm sitting here in Dublin, you're sitting in Australia, and we're talking as if we're in McKenna's. Exactly, exactly. Except I have sunburn. Okay, right. You would never, ever get things. So just keep this image in your head of four middle-aged, well-intentioned people having pastoral care towards each other as one of them goes off the rails.

15:25David McWilliams: That's the first thing. And these are friends who have known each other for 40 odd years, maybe more, right? The second thing is the power of technology. The third thing is that sometimes people lose money. And the fourth thing is, are we in a bubble like the late 1990s, early 2000s tech bubble and telecom bubble? And if so, what happens to AI? That's the nature of this podcast. Just to give you some background noise, John, it is thought that this is the biggest capital expenditure ever. But there has never been a technological period of excitement like this. So the super boosters are what they call the mega scalers, are a variety of companies, but they are in effect the magnificent seven on the American stock market.

16:18Yeah.

16:19David McWilliams: Now, those companies account for 40 % of all the wealth gains in the United States. So think about what's going on here. People who have bought those stocks feel very, very rich. They account for in the first half of this year, half of the growth rate of the United States is explicitly the money that they are spending on data centers. And what are they spending money on data centers for? It's because AI and machine learning need phenomenal amounts of data space. Now, is this the technology of future AI? It probably is, John. It has the capacity to change the way people live. So that's okay. But the key question is, what are they spending the money on, right?

17:11David McWilliams: And there's things, John, called graphic processor units, GPUs. And they are the little chips that actually propel and drive the AI. Yeah, the power behind it. Yeah, yeah. Yeah, and these are the sort of things that your brother would have been involved in making in Intel and these sort of things. Now, the fascinating thing is that right now, think about the amount of money between 300 and 400 billion. Now, you might say, well, certainly you have a proper figure on this. You can't be like a billion out or 100 billion out. The reason you're 100 billion out, it's like, oh, really? That's kind of a very strange sort of a rounding error.

17:49David McWilliams: But the reason this is the fact, John, and this is the case, is because of the fact that a lot of what the companies are doing has been masked. So let's just look at what's happening. Just before you go further there, because this is kind of comparable in the way the dot-com bubble that burst in the early 2000s. Like, what was the level of investment at that? Much smaller. Much, much, much smaller. Like, are you talking 10 % or are you talking less than that? I'm not too sure, but I know it was much smaller. And I also know it was much more limited to equity, not debt. And I explain all that. Yes.

18:29Yeah, do please.

18:30David McWilliams: So, for example, the AI boosters say, look, this is a technology that is going to be like the railroads. If it is a technology that's going to be like the railroads or.com, what we then have to build into our model of looking at this is that there's going to be a crash. because railroads and all, whether it's radio, whether it's automobiles, all these new technologies went through the roof. Tulips, exactly, up, down, up, down. At least tulips had nothing except sort of fashion value. But think about radio. Remember we talked about radio shares going through the roof? Yes, you did. Did radio change the world?

19:08David McWilliams: Yes, it did. But you go ups and downs and ups and downs. You have the crashes. So then you've got to see, okay, what is going on in the United States? What's actually going on? And so data centers are split into three costs, right? The first cost is for the chips. The second cost is the actual electricity that you need to cool these things down. Absolutely. Enormous amounts of electricity. And of course, the third cost is the actual construction cost itself. So basically what you have as a data center is a big breeze block, which has to be cooled down because the computers generate enormous amounts of heat.

19:43David McWilliams: And as they are cooled down, what you'll find is about 70 % of the expenditures on chips. And then the other bits are energy and construction costs. So that's what's happening, right? And they're building them all over the United States. And of course, what people are saying now is, hold on a second. These things are sucking in huge amounts of energy, all that sort of stuff, but they are what they are. That's what they're building, right? But the problem is the things that they're spending most money on are also very heavily accelerating technology. So every new GPU is getting faster than the one before.

20:27David McWilliams: That's the key. That's the big race between NVIDIA and Intel and all that is basically who's going to build the fastest chip, right? And what that means is if you're investing in this stuff, John, the life cycle of these chips is very short. they are in effect a perishable good. It's like buying cabbages, right? Technological cabbages. Yeah, it's like buying fucking lettuce. Like Liz Truss. Yes, exactly. So it's like buying lettuce. So it's like buying the technological industrial equivalent of a lettuce. This is the first thing that you get alarm bells ringing. So when you make one of those GPUs, when you buy them, it has got obsolescence built into it because the very competition for AI is to come up with the fastest one.

21:09Yeah, yeah, yeah.

21:10David McWilliams: First of all, the CapEx. So typically when you have capital expenditure, let's say, for example, you say, I'm going to build a railroad, a metro station, right? That metro station, that metro line, will be operable for 100 years. You know, most of the underground stations in London have been open for over 100 years. So there's a long-term, non-perishable return to the investment. If you're buying the industrial equivalent of a lettuce, In a year's time, that's going to be out of date. So therefore, these companies need to generate income very, very quickly in order to cover the costs of their investments because they're essentially buying industrial lettuces.

21:54David McWilliams: That's the first thing. Yeah. The second thing is because the energy bills are being paid for by everybody. So people in Ireland's electricity bills are going up because of data centers. Yeah. It's only going to be a matter of time that your average American voter says, hold on a second, I don't want a data center beside me because I'm paying it through my monthly electricity bills. Or even more to the point, Mac, which is beginning to happen in places like Virginia, which has apparently the highest concentration of data centers in the world, is that local towns and villages in Virginia are now having power cuts.

22:34and people are going, of course they are. Hang on a second.

22:37David McWilliams: What's going on here? I know it's the South Africanization of Virginia. Because you know in South Africa, they have brownouts. Yeah. And brownouts are when the grid just stops. Yeah. And the place goes dark. So again, another risk to this is basically natural nimbyism. People say, well, hold on a second. Not in my backyard. Yeah. Why? Not because I don't really like the hum of the data center, which hums all night. But more importantly, people are paying through their bills for the electricity that's been consumed by the large data centers. Now, the problem is utilities, as in your ESBs of this world, they love the data centers because what they're doing is they can write one contract with Meta and get paid for the electricity that might take 100 ,000 homes.

23:21David McWilliams: And 100 ,000 homes, the default is going to be much higher. So they love this. So there's a natural, this is back to the concentration idea, right? And then, of course, you have the idea that, you know, the construction of the data centers in terms of the actual building, that is long-term capex, right? But the actual essential thing is perishable. It's an industrial lettuce. So that, A, they have to figure out how they're going to make money quickly enough. Now, none of them are making money. Nobody pays for ChatGPT. Nobody. The whole world pays for it. It's all free. Grok is free. ChatGPT is free.

23:55So then you think about what else is happening.

23:59David McWilliams: And what else is happening? in AI is that AI is eating all the other tech jobs. So right now, for example, in the old days, and one of the old days, I'm talking about three or four months ago, tech jobs were quite varied. But now what you see is it's all about AI. It's writing the algorithms. It's all this sort of thing. It's basically making large language models, which are the essential core of AI, Yeah. More effective, quicker, et cetera. Right. So what you're seeing is two things. One is all the resources of universities, engineering kids or whatever, have been sucked into this bubble sector.

24:42David McWilliams: But number two, and this is really important, John, is when there is a concentration of enthusiasm and excitement in a certain sector, number one, and when there's a concentration of finance coming from a certain sector to that sector number two, all other sectors get elbowed out and don't get the investment they need. This is back to the Dutch disease idea, right? So AI is the Dutch disease of modern America. So for example, if you're a manufacturing, and it's weird because the Americans are obsessed now with manufacturing and tariffs and all that sort of stuff. If you're making, I don't know, this microphone that I'm talking to, right?

25:18David McWilliams: It's going going to be much harder for a microphone maker to get access to finance when all the banks and financing companies and private equity and private credit companies really just want to write a check to meta or google because that's where all the action is yeah so what you see what you see is a the amount of scientific endeavor that universities are producing is going into one sector but b all the other sectors are being elbowed out which is making america more fragile not less fragile because we all know that what makes economies robust is diversity. I mean, lots of people doing lots of different things.

25:53David McWilliams: So this is where we're going with this stuff, John. But what does that say about the investors themselves? You know, I don't want to go down a rabbit hole here, but it is this FOMO kind of mentality. And I always thought, well, when I was growing up and I talked to people who were investors and working in banks and investment houses and all the rest, that you kind of felt that those guys are analyzing and they're this, that and all this, and they know what they're talking about, don't they? No, they're not, John. I worked as one of them. Jesus. Well, I know you don't know what you're talking about.

26:31David McWilliams: Exactly. Speaking of rabbit holes, I've just got out of one, as I just told you. That's a K-hole. Anyway, come back. This is a really good question because it leads us on to the other thing, which is the changing nature of finance. Right, okay. So in the old days, there used to be banks, many, many banks with people who had relationships with many sectors of the economy. So you took a very big bank, I'm looking at the window here, ANZ, the big Australian bank, right? Australian New Zealand bank, they would have had relationships with construction, with finance, with technology, with agriculture, with aviation, and they would have had units who'd be lending to airplane manufacturers or whatever, right?

27:10David McWilliams: Yeah. In recent years, we've seen the emergence of something called private equity and private credit, which is a sort of superannuated finance house. And these are beginning to dominate in the United States. So as the banks after 2008 retrenched and basically licked their wounds and realized that, you know what, it is not prudential for us to be led into every Tom, Dick and Harry, that gap was filled by private equity. Now, private equity is a bizarre concentration of risk because private equity prefers to have less clients and take bigger bets than have more clients and take smaller bets. Why is that?

28:00David McWilliams: Because the partners of private equity are really greedy. and they don't want to build large banking style firms with all the costs. What they want is they want to make out like bandits and earn$20 million a year. Now what that requires is one big bet that comes right. So we're seeing a concentration of money in private equity and therefore they're much happier getting into bed with Meta Because what they convince themselves is, oh, well, the partner on the other side is Meta and they're really secure. Why? Well, because Meta, Facebook is one of the biggest companies in the world. They have loads of money, et cetera, et cetera.

28:43David McWilliams: So that's all fine. But what it's doing is something very interesting. So Meta and Google, rather than spend their own money on the data centers that they are building to create the AI future, have decided to partner up with private equity. They're splitting this, okay? Now this then comes into a very strange world because what is in effect a bet on AI

29:16David McWilliams: becomes turned into a bet on data centers, okay? And what is a bet on data centers is being sold to investors as a property play. So you will find that even in Ireland, brokers are going around Ireland saying to their top clients, look, why would you bother investing in a hotel in Dublin or a residential office? You can invest in the data center. And the end client is Google. And Google can never go bust. Yeah. It makes a certain amount of sense, in fairness. It does. But it means that retail property investors now and governments who are actually behind all this and involved in all this are no longer property investors.

30:03David McWilliams: They are AI speculators. Right. Because what they're buying is the notion that the microchips will change the world, even though they're not being told that the microchips are industrial lettuce. And the reason they're not being told it's industrial lettuce is nobody can admit this. Why? Because the gap between revenue and costs right now for the mega scalers is enormous. And by that, I mean, nobody's paying for AI yet. Yet the cost is real. So right at the moment, the investors in all the AI companies and the NVIDIAs and the, you know, whatever they're called, OpenAI and all these companies, they are subsidizing a dream, which is that AI will change the world.

30:52David McWilliams: Now, this is where you have to really start worrying about bubbles, reality, differences, old-fashioned things called profit, John. You might have heard that thing. It's quite a good thing. Yeah, yeah, yeah. I'm not familiar with that, believe me. But at the moment, with all the FOMO in the investors market and all the AI that seems to be going at 90 to the dozen, why are you saying that it is a bubble that could burst anytime soon? Well, let's discuss this after the break, Sean. Indeed.

31:58We'll be right back.

32:07isn't one size fits all. That's why drivers have enjoyed Progressive's Name Your Price tool for years now. With the Name Your Price tool, you tell them what you want to pay and they'll show you options that fit your budget. So whether you're picking out your first policy or just looking for something that works better for you and your family, they make it easy to see your options. Visit progressive.com. Find a rate that works for you with the Name Your Price tool. Progressive Casualty Insurance Company and Affiliates. Price and coverage match limited by state law.

32:39David McWilliams: So we're back in the room, dear listener, right? So the question is, why is it a bubble? Because I believe the boosters. I believe Sam Altman. I believe Mark Zuckerberg. They all tell me that this is going to be a transformational technology that will change the world. And this will be the equivalent, as Jeff Bezos says, of an industrial bubble, not unlike railroads. Well, I will remind listeners that railroads caused six panics in the United States. Six panics? It caused panics, financial panics, over a period of about six decades. Right. Electricity similarly, telecom similarly, automobiles similarly, radios similarly.

33:23David McWilliams: Why? Because technology promises the speculator something that the investor doesn't want. The speculator wants super profits very, very quickly. The investor wants moderate profits over a long time. So the difference between the speculator and the investor as creatures in the financial ecosystem is the speculator is always attracted by technological breakthrough because the promise of technological breakthrough is absolutely transformative profits. The investor, on the other hand, is not that interested in technological breakthroughs. But the speculator creates the background noise for bubbles because the speculator is, in effect, a proselytizer.

34:10David McWilliams: It's a recruiter. The speculator recruits people into his worldview. That's your FOMO idea. And what you find always in all these great technological transformations is that they spike up and down. There's bubbles up and down. We get too excited. Then we lose money. Then people get depressed. Then they sell too quickly. etc. It's the great Minsky-Kindelberger, dear listener, if you're interested in this, Minsky and Kindelberger, two great economists, cycles of booms and busts. So we're in one of those, which is it's more psychology than economics, right? It's group psychology. It's all about people doing things because other people are doing things.

34:50David McWilliams: So that's why I think we're in a very clear AI bubble. But the interesting thing is Zuckerberg thinks the same thing. If he's a student of economic history, you'll know that this happens. But what is interesting, though, in all of this, just to bring it back to reality, is that it's very easy talking about finance and bubbles and tech and how marvelous it is and could be and all the rest. But it's not taking into consideration your average Joe. So when you say the bubble bursts, it's real people that actually suffer from that. Exactly. Because of finance. Because finance, money is social. Yeah.

35:28David McWilliams: Because you don't realize that if this bursts, the risk appetite of Bank of Ireland is going to be profoundly negatively affected. And therefore, the amount of money available to mortgages in Ireland, for example, will diminish. Because there'll be a large sucking sound in the global financial markets of liquidity drying up. and liquidity is in effect easy money. When things are going up and everybody looks to be making money, liquidity is ample. This is because commercial banks create liquidity based largely on today's news. When liquidity disappears, that liquidity disappears across the board.

36:11David McWilliams: And that's how the average Joe gets affected. That's the first thing. The second thing is think about Ireland. Ireland is getting all this incredible corporate tax revenue from the Magnificent Seven. Yeah. If their balance sheets implode, our corporate tax rates implode. And not only that, spending on houses and schools and roads around the country stops. Yeah. That's how the average... This is why it's so important because we in Ireland... I'm in Australia here. It doesn't matter. we're all little cogs in a big machine. And the biggest cog in the machine is the American capital market. And right now, the American capital market has gone off on a sort of speculative mania around AI.

37:01David McWilliams: And that's why it's important to try and decipher it, disentangle it, and to understand who's benefiting, who wants this, who doesn't want it, who's going to pay if it all comes tumbling down. Okay, so let me ask you then that, you know, we've seen all this before with the tulip mania and the railways and the dot-com bubble and all the rest. We've seen this. We've seen this bubble burst cycle. America seems to be the only country in the world that is not taking any sort of steps like antitrust steps. and is taking an antitrust kind of regulation, is that the way to avoid a burst of a significant magnitude that it takes the whole shit down?

37:52David McWilliams: I think what we have, I'm going to answer that question. I'm going to conclude the podcast because we could do a number of podcasts on this, but I just want listeners to be aware of what's going on and to be aware of what's in the data centers, where the spending is going. Sure. All that sort of thing. First of all, we have a race against time. All the money that's been spent on AI is a speculative bet that one company or two companies will emerge as the dominant ones. Yeah. Because by definition, somebody has to lose. That's the first thing. The second thing is the race against time is because it's all about the attitude that investors have to risk.

38:33At the moment, none of these companies are making money.

38:37David McWilliams: Nobody's paying for AI. Nobody's paying for chat GPT. So the revenue is zero, but the costs are enormous, right? So what these companies are hoping for is the risk appetite of investors will be sufficiently long so the investors don't panic because it's the investors who are buying the stocks of these companies that are actually financing it and subsidizing the dream of the executives. And this is like the Amazon model. This is like the Amazon model. It takes time. That didn't make money for years and years and years till they gobbled up all the competitors or potential competitors. Exactly. So once they gobble up competitors, it means that the actual people who own those competitors lose money.

39:17David McWilliams: So my point is there's a race against time going on. The third thing in the United States is really important to appreciate, which is that the people involved, the engineers, the scientists, that technologists involved in AI actually don't care about the financial fallout of the bursting of this bubble or the social fallout of the financial fallout. And I know this because I was given a book last year, which is published by Stripe Press, and this is important, called Boom, Bubbles and the End of Stagnation. Right. Two guys called Byrne Hobart and Tobias Huber. It's a very good book. It's well worth reading.

40:03David McWilliams: What is fascinating is they make the case for financial recklessness because bubbles are in effect a nice word for being financially hyper reckless. And what they say is that only via hyper reckless finance and money can you create the enthusiasm and the excitement for a massive scientific breakthrough. In the book, they make the point of the Apollo project, as I said earlier on, or the Manhattan project, as I said earlier on, where there's three things common in all these projects, right? One is that it's a very small group of people who are involved. Two is they have access to vast financing.

40:56David McWilliams: And three, the accounting standards governing the investments are very lax, right? Now, this is really important to get our heads around. Why? Because they're driven by propelling the economy onto a higher technological plane. They don't care about finance. So automatically what you have is Wall Street, the financial guys, and Silicon Valley, the guys they are piggying back on and financing, have actually two different end objectives here. Wall Street want to make a profit. Silicon Valley want to change the world. These are two very, very different impetuses. And once you realize that Stripe Press is publishing that book, that in itself is interesting because they are technologists who want to change the world.

41:47David McWilliams: Yeah. And then, of course, there's a third leg to this stool, which is democracy, which is politics. Because Wall Street wants to make a profit. Silicon Valley wants to change the world. But democracies want to look after their people. Yeah. And if these things implode, so if there is a massive crash, if there's a huge speculative reversal, right, Wall Street starts losing money. The data centers start losing money. All those data centers in West Virginia then become redundant. What you find is the jobs associated with the growth rate in the United States, which has been driven by AI, disappear.

42:29David McWilliams: If the jobs disappear and the effervescence and the good times disappear, we're facing to an electoral cycle. President Trump, Mark III, because I suspect he will run again, will be absolutely cognizant of this. So because this sector is so unbelievably important to the United States economy, if it unravels, the growth rate unravels, the unemployment rate rises in the United States, house prices fall in the United States, bank lending falls in the United States. Why? Because the tolerance for risk disappears, liquidity dries up, The United States goes into an AI-inspired recession just when Donny J needs a boom to get him over the line.

43:19David McWilliams: And that is the framework and the timeframe that I would like listeners to put into their back pocket and ruminate on as they're on the dark or the metro or slightly or going for a walk or wherever you're listening. Or on the beach getting sunburned. Or in the back. no it was even worse John in the back of a boat down the Yarra River realising there was a reason that only me and a whole bunch of Indians were at the front anyway I'll tell you next week I'll be coming home next week excellent stay away from those jellies just bring them home never again never again it was the it was the chocolate favourite mushroom cake that just tipped the whole thing over the edge me and the bikers me and the hell's ancients see you later boss Good luck.

44:37up to 70 % off during Wayfair's 4th of July clearance. Score huge deals on outdoor furniture, area rugs, and more. We're talking thousands of products for every style and budget. Plus, surprise flash deals July 6th. Don't wait. Shop Wayfair's 4th of July clearance now through July 6th at Wayfair.com. Wayfair, every style, every home.