#305 -The AI Bubble Is Coming For Your Super

Everyone's talking about the AI bubble. 54% of global fund managers think we're in one. The International Monetary Fund is warning about it. And Michael Burry, the guy who shorted the housing market in 2008, just shorted a bunch of AI companies in the last few weeks. And yet everyone's still piling money into this space like it's a sure thing. But here's what nobody's telling you. The AI bubble isn't coming. It's

already here. And the only question left is how bad the crash is going to be and whether you're going to protect yourself before it happens. And if you're an Australian investor, if you have superannuation in international funds, you're more exposed to this than you think. And in this video, I'm going to show you exactly why the AI bubble is unsustainable. I'm going to show you the circular deals that are propping up the entire industry. And most importantly, I'm going to show you

what this means for you as an Australian investor. I'm Lloyd James Ross, seven-figure investor and entrepreneur, and I've helped thousands of business owners and professionals turn financial stress into success. If you're stuck in old money habits, overwhelmed by investing, or unsure where to start, this is for you. I'll give you the mindset and strategies to take control, grow your wealth, and achieve financial freedom. It's time to make your money work for you. Let's

start with this, the scale of the bubble. We'll start with

the numbers. Because once you see the scale of what's actually happening, you'll understand why it can't last. So OpenAI, the company behind ChatGPT, was recently valued at about $500 billion. People are saying when it lists it's gonna be a trillion, but $500 billion. Now for context, that's more than Coca-Cola, more than McDonald's, more than even Pfizer. And how much revenue does it currently produce? Well, it's generating about $10 billion a year. How

much are they losing? More than that. They're burning through like eight and a half billion dollars in losses, just in losses annually. So we have a company valued at 500 billion that's losing money. And not just losing a little bit of money, it's losing billions of dollars every single year. But don't worry, it gets worse. OpenAI has committed to 1.5 trillion in AI deals. That's trillion with a T. They've signed $500 billion

deals with Project Stargate. That's to build all the data centers in the US. A $300 billion deal with Oracle, specifically to buy all the compute power. And another $500 billion with NVIDIA to buy the chips. And then another $300 billion with AMD. That's 1.5 trillion in commitments for a company that generates 10 billion in revenue. So let me put that in perspective. If open AI used a hundred percent of its revenue, every single dollar it makes to pay for the deals, it would take 150 years to

fulfill those commitments. It's not a business model, it's almost like a Ponzi scheme, right? And

OpenAI is not alone in this. McKinsey did a recent study and they estimate that the AI industry will need to spend, check this out, will need to spend $7 trillion in capital expenditures over the next five years just to meet the demand. $7 trillion! So to put that in context, America's total capital expenditure across all industries in 2024, last year, was 8.3 trillion. So the AI industry, just one industry, is planning to spend nearly as much as the entire US economy spends on

everything. And here's the thing, most of these companies aren't making money. So there's a circular monetary problem happening here. You might be thinking, okay, well, aren't these people smart who are running these companies? They wouldn't be investing in this stuff if it didn't work, right? And I understand that, but here's where the circular money problem arises. You've seen this chart floating around probably, it's floating around the internet, and

it shows the financial relationships between all the major AI companies. And when you look at it, you realize something very disturbing. They're all just paying each other. So here's how it works. NVIDIA invests $100 billion in open AI because NVIDIA is one of the largest chip providers, right? In fact, it's the largest and it's a $4 trillion company, a $5 trillion company now. And it's the largest company in the world. And here's the thing, it's

got what's called vendor financing. And vendor financing is where they actually allow their biggest customers, and OpenAI is one of their largest customers, $100 billion, to actually buy the chips off them with borrowed money from them. So this happened in the dot-com bubble too, and it's now starting to happen again. So in return to that, OpenAI agrees to buy NVIDIA's chips. And so they're kind of incestuous, they're

in bed together, right? And so NVIDIA also invests in the CoreWeave, which is an AI data center, and they agreed to backstop all their services if they can't find customers. So CoreWeave then

sells computing power to OpenAI. And AMD gives OpenAI warrants for shares in return for OpenAI's You know commitment to buying the AMD chips so that announcement alone sends the AM stock ticker higher Which then helps pay open AI for the deal that makes sense because they're using warrants So open AI announces a 300 billion dollar deal with Oracle for data center capacity Oracle turns around and buys chips from media Who's invested in

open AI see the problem? It's a circle. It's circular and NVIDIA is investing in its own customers so they can then buy NVIDIA chips. And then NVIDIA is reporting those chips as sales as revenue, even though they're effectively funding the purchase themselves. And that is the vendor financing conundrum. And it's exactly what happened right at the end of the dot-com bubble incidentally. So the company that was doing that was say

Nortel for example. They started lending money to their customers so they could buy their products. And it made their revenue look great. It made their profits look great. And then the whole thing collapsed. Because here's the truth. If you're investing in your customers to give them the cash they need to buy your stuff, you're not actually making any profit. You're just moving money in a circle. It's

a loop. And when the music stops, and it will, when investors stop putting money into the space, the whole thing falls apart. And then there's the demand

problem. And let's just say I'm wrong. Let's say these circular deals aren't even a problem. Let's say these companies can actually deliver on what they're promising. There's still one massive issue, demand. Specifically, when Bain and company did a study, they estimated that AI will need like $2 trillion in annual revenue by 2030 to be profitable. Think about that. In four years, they need 2 trillion in

annual revenue to be profitable, 2 trillion. That's more than the 2024 revenue of Microsoft, Meta, Alphabet, Amazon, Apple, and NVIDIA all combined. That's five times more than the entire software as a service market. And there's no path to that kind of revenue. Yeah, so 80% of companies are using AI in some sort of capacity, right? So ChatGPT broke the records and it's the fastest growing app ever. OpenAI now has like 800 million active users, but only

6% of those users are actually paying for the service. And 61% of the companies using AI say it hasn't had any real impact on their earnings. So we have hundreds of millions of people using AI for free effectively. And the companies that are paying for it aren't seeing a return on the investment. And Citigroup estimates that AI revenue will reach like $780 billion by 2030. It's very impressive growth. That's like

80% a year. It's incredible, right? It's an industrial bubble. But it's still less than half of what Bain said that they needed for the industry just to break even. That's assuming everything goes perfectly. That's assuming AI companies deliver on their promises. That's assuming that demand keeps growing at the same pace. But what if it doesn't? What if we're in this Metaverse 2.0? Remember when

Zuck started spending all the cash flow from Meta on the Metaverse? And remember everyone said it was the future, et cetera? I can't remember that. And Meta stock just collapsed because it was just wasted money. Because where is the Metaverse now? It's like, remember when people were buying land in the Metaverse? What happened to them? Anyway, AI is certainly more promising than the metaverse, but the same principle applies. Just because a technology is exciting doesn't mean

it's profitable, doesn't mean it's a good investment. We've seen this before with railroads in the 1800s, bubble, industrial bubble. We saw it then with cars, industrial bubble. We saw it then with the internet, industrial bubble. If you look at these industrial bubbles over the course of history, The railroads, there was like tons of railroad companies. And now there's only like a handful. There's only like a handful. There's only like eight big class one railroads in the Western markets,

really. And so you think of all those railroads, there's only a handful. Then you look at cars. In fact, in America, there was 200 car manufacturers in the United States at the beginning of the car boom. Now there's like Ford and GM, two, there's two, two. The internet, remember all the internet companies weren't making money? They're gone, gone. Pets.com, toys.com, you know, all gone. There's only a handful, Google, Amazon, Microsoft. Like

there's just a handful now. So if you're looking at this as an industrial bubble, of course, there's only gonna be a handful of AI companies left. So there's gonna be a huge fallout, right? So here's the thing with, if you're listening from Australia, cause that's where I'm based right now. Here's

what nobody's talking about here in our country. What does it mean for us? What does it mean for people who are, you know, it's in America, right? Open AI doesn't really matter. So what happens to Australian investors when the bubble pops? Let me. Well, it's not good, particularly lately, because a lot of superannuation funds are gonna start putting a lot of money into international shares. But according to the ASX, the

Australian investors have poured billions into AI as well. NVIDIA, Microsoft, AMD, these are some of the most heavily traded stocks, even on Australian exchanges, because we can also, on our ASX, have index funds like, you know, from Vanguard and BlackRock, they're listed on our exchange. So you can actually have huge exposure to these companies via those types of ETFs and

funds, and that is actually happening. So it's not just retail investors, it's Australian super funds, it's your retirement savings, it's all that stuff. They're heavily exposed to AI. If you look at some ETFs, That a lot of Aussies are exposed to like I would say 30 to 40 percent of the holdings are all exposed to AI It's crazy, you know it's wild and so most people are going to be exposed to this crash, right? Or

this adjustment in the bubble. Just quickly, if you're ready to take control of your finances but feel stuck on where to start, I have a solution. My book, Money Bias Happiness, simplifies investing and wealth building with practical steps to help you achieve financial peace. Get your copy via the link in the show notes and let's get your money working for you. Now back to the episode. So when it pops, it will definitely hit retirement savings. But there's an

even bigger risk if you're in Australia too, because you've got currency risk. So most of these investments

are in US dollars. So when the bubble pops, you're not just losing money on the stocks, you may also have some sort of currency conversion too. So yeah, just be careful of that. Because when the US market crashes, typically the Aussie dollar strengthens. And so that amplifies your losses as well. With any hope, people run to the US dollar and that doesn't happen. So you could potentially get hit twice, right? One with the stock price, and

then one on the currency. And there's the economic impact of that. And people will often say, well, how does that impact our economy? Well, remember when housing collapsed in America? Boom, a couple of years later, we got smashed. Because the old saying goes, when America catches a cold, when America sneezes, Australia catches a cold. You've probably heard that before. So Australia's economy is definitely dependent on resources, right? We've got iron ore, coal, natural gas.

These are our biggest exports. And guess who buys them? China. And

guess who else buys them? USA. And they buy them for manufacturing, infrastructure, technology, and development. And so when the AI bubble pops, the global thirst for these commodities will actually dry up too. Now, I don't think that's going to happen in the short term. But long term, at some point, there will be an adjustment to these revenue projections, the amount of capital being

put into these things, and the ROI that's not coming back. and that's gonna definitely have a knock-on impact to us as a country, to our resource sector, et cetera, and then it'll start to then have a knock-on effect in jobs, right? So it means the Australian economy then will start to contract as well. So it wouldn't surprise me to see a recession after a global AI bubble pop. So everybody's pretty much exposed to

the risk, okay? So it's not just gonna hurt tech investors, I think it's gonna hurt everyone. Now, to what extent, we don't know, but there's definitely concentration risk because Everyone's talking about AI. It's just like the thing. Yeah And I think it's like 35 to there about companies account for like 90 Yeah,

almost a hundred percent of the AI token spending In fact, there's two companies that are actually accounting for almost 50% of it, right? NVIDIA's chips, of course, and then other companies involved in the AI itself. They account for like, I think it's 36% of the S&P 500's growth. This is wild, okay? I would say the most concentrated

S&P 500 in recent memory. So when you're buying the S&P 500 as an example, you're buying 500 companies, but by the way, most of the gains are from The AI companies which account for 40% of the fund. Okay, so it's just something to understand, right? So if one of these companies fails, then it sends a bit of a shockwave through the entire market, right? And then OpenAI is basically at the center of it all. Because OpenAI's deal announcements alone, right? They've added hundreds of

billions of dollars to the market values of Broadcom and Video and AMD. So if OpenAI pulls back, if they can't deliver on their promises, and right now Alphabet is cranking. It's taken so much market share from OpenAI. Look at the chart. It's dominating because Google has a distribution across its powerful

businesses to implement AI faster than OpenAI. OpenAI is pure. It's very much its own business, and it can get absolutely crushed by people like Alphabet, who have just got these wide modes and distribution channels that they don't access. In fact, right now, Google is one of the only ones. In fact, Google and Alibaba are the only full-tech stack AI, which means this. They own the chips, they own the data centers, and they own the distribution. the actual AI models. So

they're the only full stack. And in fact, too, they've got contracts on nuclear power. So if you go right upstream, nuclear power contracts for getting power, then you've got the actual chips themselves, then you've got the data centers, then you've got the models. And that's a full stack AI for vertically integrated business model. Those who've got a fully vertically integrated business model not only control the model themselves, they make better margins, but they also can squeeze

out these other ones at a higher cost. It's very much like the oil industry in that sense. You get companies like that versus an open AI and they're already crushing open AI. This is what's happening in real time. What's happening too is in

the prices out there, the P ratios, the price that you're paying for these companies in relation to their earnings, they're getting quite high, which means there's no room for error. So there's no room for disappointing results. If they don't stack up, it doesn't happen, they don't deliver, there's gonna be massive price adjustments. You're gonna see a huge, there could be a big bear market, 20, 30% collapse in the stock market for sure, as

these prices start to adjust based on the earnings that they can actually deliver on. So you look at the peer ratio of the S&P 500 now, it's actually at 30 times earnings, which historically is very high, very high. It's not quite as high as the dot-com bubble. In the dot-com bubble, it was 46 times earnings. Flipping very high, right? So we're not quite at dot-com

bubble levels yet, but we're getting close. And here's the difference. During the dot-com bubble, 36% of the tech companies were actually losing money. They're just like URLs, it was wild. I was only 17, I still remember. Today, it's only 18%, right? So on the surface, things look a lot better. The financials are definitely stronger. The companies look more stable. They are more stable, but that doesn't account for the private companies like

OpenAI and Anthropic and so forth that are completely unprofitable. They're burning through billions of dollars in cash. And it doesn't account for the circular money problem with the vendor financing that's happening with Nvidia, or the demand problem, or the infrastructure bottlenecks like power, right? Which is a huge constraint. So why this bubble, this industrial bubble, which it is, is different from the dot-com bubble, which is also an industrial bubble from the internet. It

is different. And I know you're probably thinking, Lloyd, why are you being so negative, right? Why are you bullish on AI? I love AI, it's great, but there's definitely a bubble. The internet was a bubble too, and it's changed the world, it's amazing. So financial bubbles are really bad for people. Like the 2008 housing bubble was a financial bubble. It was FOMO by the consumer of buying things worth way more than what it costs, and it

was FOMO. That was very bad because they were selling bonds, anyway, long story. Not ideal. Financial bubbles suck. 1929 was a financial bubble, and so was the 2008 financial crash. So financial bubbles, extremely systemically bad for us. They can cause depressions. However, industrial bubbles are really good for us as consumers. It advances our lifestyle massively. But it's not so good for investors if they own pieces

of these companies that aren't making the money they say they are, right? That's the issue.

So, the internet changed the world, railroads changed the world, cars changed the world, AI is changing the world. But it also, the internet, dot-com bubble also led to an 80% collapse of the NASDAQ If you look at the NASDAQ in 2000, 1999, 2000, 80% collapse. In fact, it collapsed over a couple of years. But then it, listen to this, it took 14 years to recover. Be aware of that. It can happen again. So thousands of .com startups obviously went broke. Investors lost trillions. It took a long time

to recover. But again, a few companies did survive. So yeah, AI is changing the world, but it doesn't mean you're necessarily going to make money from it. And this is why Warren Buffett doesn't invest heavily into fast-changing industries, because it's not how you make money. It's how you definitely have a better life.

Okay, so there's a few differences between the dot-com bubble and the AI bubble back then interest rates were rising The Federal Reserve was actually tightening monetary policy right now. They're actually gonna bet to the opposite They're gonna probably drop rates because inflation's come right down. Okay, so there's a huge difference. We're gonna see rate cuts which means the bubble's got more room to

run. If rates are cut, oh my God, we're gonna see this thing flip and boom, and I think rates are gonna get cut, because Trump right now is like, he's hammering the Fed chairman to start cutting rates, right? So the timing's a lot different. During the dot-com bubble, Alan Greenspan was the Federal Reserve Chair at the time, right? And he said, oh, there's a lot of irrational exuberance, which there was in 1996, I think he said

that. And then four years later, bang, the bubble pops. So it takes a while, right? So even if we are in a massive bubble, it could take a few years. And especially if they drop rates, it definitely will. And it could also pop tomorrow. I mean, we just don't really know. But we do know for certain that right now there's way too much capital flowing into AI based on the expected returns of

that capital. Now, if that happens, it's very capital intensive and very return light, which means at some point it'll adjust. And when it adjusts, the market will go through some sort of you know, bear market, okay? So what do you do right now? What do you do? First, if you're invested in AI stocks specifically, you need to take a hard look at your portfolio. I'm

not saying sell everything, I'm not saying don't buy AI stocks, but I'm just saying you need to understand your own risks. So if you own NVIDIA, AMD, Microsoft, one of these companies that are not only heavily exposed to AI, but also at a very high price based on the capital expenditure, Then you've got to be very careful. Now, there are companies out there like Google, for example, got multiple verticals. It's got YouTube, cloud, it's a cloud leader. It's got its own TPUs now. It's

a vast empire. People often joke, Google is the internet. And so you've got these companies that are earning tons of profit, got a hundred billion in cash. There are certainly opportunities out there where it makes a lot of sense, where companies are really going to they're gonna become even more powerful with the likes of AI. So there are some great opportunities out there, as long as the valuations are reasonable. And that just means are you getting, are they gonna be able to generate

profits based on their spending? That's all that matters, okay? So if they can't, they'll fall hard, right? Now the other part that you can do is make sure you're diverse. So don't put all your eggs in one AI basket, okay? You don't wanna chase the hype. I see some portfolios in X and they're just all AI stocks. I'm like, these guys are gonna

get killed, right? So make sure you're diversified across sectors. Perhaps you'll be in like, for example, I'm in railroad, I'm in oil, I'm in healthcare, I'm in certainly technology, but I'm in China tech, so it's a little bit different to US tech, and I'm slightly a little bit in US tech, and then a little bit of cash and some other things. So you have to be diversified. That's very important, right? You don't need to just be all stacked in AI because you

have no sector diversification. Everybody knows to be a little bit diversified. And then understanding that owning shares in companies that survive the bubbles, that's a more astute way to invest long term, right? And that's where I'm putting my money because the companies that I'm investing in, I know that even through a bubble burst, they'll come out stronger. And that's what you want to do. You want to invest in companies that will do well either

which way, okay? And you've got to be very mindful of price. You don't want to time the market. That's not a good, sensible strategy. What

I've done lately is I've got this thing called the 100-year portfolio that I'm building, my own personal portfolio. And I've renamed it that because I'm like, am I buying companies that have not only withstood the last 100 years, that they can do it again, in products and areas that are impossible to compete with? And it's just changed my entire approach And I'm just so much more excited about that because I don't have to worry about timing, disruption,

anything like that. And some companies in the AI race, they do tick a lot of those boxes. So you don't wanna make emotional, hypey decisions around AI, okay? Just protect yourself, diversify, focus on the fundamentals, okay? If you don't know what fundamentals are, go get my book, Money Buys Happiness, on the second page. There's a little cool little course that you can do which will explain all that stuff, right? Moneybuyshappinessbook.com, go

check it out. Because the worst thing you can do is not be educated and do nothing. That's the worst thing, right? Is not understand what's happening, right? So I'm not here to tell you that it's all gonna be bad and it's gonna, you know, it's a huge fallout and we're gonna go through a massive recession. I'm not saying that. I'm saying that AI is changing the world. I'm excited about it. Some insane things are coming, but it

may hurt you as an investor. When Mark Zuckerberg gets on the news and he tells Trump, we're going to throw 500 billion, they just throw these numbers around like nothing. they have to generate a return from the capital. Otherwise, there's going to be capital destruction. And that's what's going to lead to the AI bubble popping, right? Especially if this circular money thing continues, which is why Michael Burry's put

a short bet on these companies, because he knows this too. It's just a matter of time, right? So when it pops, a lot of people are going to lose a lot of money. Don't be one of them. Protect your wealth, diversify, focus on fundamentals. And if you found this video helpful, make sure you hit the subscribe button because I'm going to keep tracking this story. And when the bubble pops, you're going to want to know what to do afterwards.

There's going to be a fallout. Like, what do I do now? We're going to be doing lots of episodes on this, right? So thanks for watching. Stay safe out there in the AI bubble world, and I'll see you in the next episode. Thanks for listening to Money Grows on Trees. If you enjoyed the episode, leave a five-star review on Apple Podcasts and Spotify and subscribe to us on

YouTube so you never miss an episode. And if you're serious about building wealth, make sure to check out the links in the show notes and follow me on all social media platforms at LloydJamesRoss for