Dive into the critical analysis of the current artificial intelligence stock boom and its staggering downside risks. In this episode, Patrick Boyle examines the intense sector concentration driving global markets, revealing how ordinary investors are heavily exposed to the AI trade through hidden avenues like small caps and value funds. The video features staggering wealth destruction estimates running into the tens of trillions of dollars from elite economists like Dean Baker, Gita Gopinath, and consultants at Oliver Wyman. It also explores historical parallels to the great railway mania and dot-com bubbles, demonstrating why boring, unexciting diversification remains the ultimate port in a storm.#PatrickBoyle#AICrash#StockMarket Bubble #FinancialNews#Macroeconomics#Diversification#ValueInvesting#TechStocks#MarketConcentration#DownsideRisk
Video Summary
AI Generated
This video analyzes the systemic risks of the AI trade, arguing that AI exposure has permeated the entire economy, rendering traditional diversification ineffective. It warns that a potential bubble burst could lead to a massive loss of household wealth and a severe recession, emphasizing that transformative technology does not always guarantee a safe or profitable investment.
Back in May, Micron and SK Hynix, two memory chip companies, produced 17% of the entire
global stock market's return for the month.
Not 17% of the chip sector, 17% of everything, every listed company in every country added
together.
That's from a report by Akkadian Asset Management.
The global market they are measuring is the MSCI All-Country World Index, which holds
thousands of stocks across dozens of countries, and Micron and SK Hynix are about 1% of it.
So, for every dollar the global market rose in May, about 17 cents of it came from just
those two stocks.
This raises the awkward question, if two chip stocks that most people on the street have
never heard of, it's not like it's Apple and Google, can haul the entire planet's
equity markets upwards, what happens to your portfolio on the way back down?
So today we're going to look at how exposed ordinary investors are to the AI trade, why
the usual places to hide may not be working anymore, and roughly how much money would
disappear if the whole thing fell apart.
Now let me clarify something up front as we're going to spend a while looking at what happens
if this all goes wrong.
There's no consensus that it is going to all go wrong.
Plenty of smart people think that we're standing at the start of an enormous boom, that AI
is a real technology that's going to reshape the economy and eventually more than justify
the money being poured into it.
And they might well be right.
In fact, you have to assume that most of the people involved believe exactly that, because
it's the only thing that explains what they're doing.
This is why investors are putting money in at these valuations.
It's why the tech CEOs are committing hundreds of billions of dollars to data centers.
No one spends like this unless they're convinced that the payoff is going to be huge.
The optimists, to be really clear, aren't a fringe.
They are the majority, and their optimism is the reason all of this exists in the first
place.
So, I'm not here to tell you the crash is coming.
Nobody knows that, and anyone who says that they do is guessing.
In fact, later on we'll look at some of the confident but wrong predictions from the past.
What this video is about is trying to estimate the downside risk.
If the bulls are right, everything works out fine, and we'll all look back and wonder what
all the fuss was about.
But if the bears are right, even if some of this turns out to be a bubble, it's worth
looking at the estimates of how much is actually at stake, particularly when the parties still
aren't.
So, diversification, it's the one thing in finance everybody agrees is a good idea.
Investors treat the world like a magical incantation, one which can protect them from market crashes,
inflation and poor decision making.
You spread your money across lots of different things so that when one of them blows up,
the others keep your return steady.
It's the closest the industry has to a free lunch, and for most of history it has worked.
A couple of years ago, the big worry amongst those who worry for a living was sector concentration
in the S&P 500.
The worry that seven companies, the Magnificent Seven, had grown so large that the index everyone
thought of as the American stock market was really just seven tech stocks in a trenchcoat.
The concern has since grown.
It's no longer seven stocks that we have to worry about, but a whole sector, and a sector
which famously doesn't stay in its lane.
AI started out looking like a handful of tech giants, and it's worked its way into nearly
everything, and I don't just mean the companies pretending to use it.
The obvious ones are the chipmakers, but it's also the utilities as data centres need huge
amounts of power, so a company whose job it is to keep the lights on in Ohio is now
partially an AI stock, or at least its valuation is tied to the idea that data centres will
soon become huge customers.
It's real estate too, because someone has to own the warehouses full of servers, and
it's construction, as there's a huge boom in data centre construction across the United
States.
The AI trade now employs electricians, and lots of them.
Then there's the money that the boom has already made.
A few weeks ago, SpaceX went public, turning around 4,400 of its employees into millionaires
overnight, including 400 of them who are now worth more than $100 million each.
The people whose job it is to sell things to the rich notice this immediately.
Estate agents in California and Texas reported a wave of inquiries about new homes.
Private jet firms report extra business from those who wanted to celebrate the IPO with
a trip somewhere.
And apparently, the single most popular purchase after an event like this is a luxury watch,
because as one watch dealer explained it, the share certificate sits in a brokerage
account where nobody can see it, but the watch goes on your wrist.
Which is a nice way of describing the urge to let strangers know how well everything's
been going for you.
The point is that the wealth doesn't stay with the AI workers.
It disperses out into the whole economy, to estate agents, watch dealers, pilots, interior
designers, the people who install the wine cellar, and the people who stock it.
All of them now care about the AI trade, whether they put it this way or not.
Your florist may have a position in NVIDIA, she just doesn't call it that.
SpaceX, of course, was just the opening act.
Anthropic is expected to go public in October at a valuation of between $1 trillion and
$2 trillion, and OpenAI is in this queue too.
Employees don't all have to wait for the IPO either.
According to the FT, OpenAI recently completed a near $7 billion tender offer
buying back shares from employees, so the watches, the jet charters, and the new homes
near the office don't have to wait for the bell to ring at NASDAQ.
The money is already spilling out into the economy as we speak, one $7 billion buyback
at a time.
Which means the sensible question is, can you get out of the way of AI, or are we all
stuck in a possible boom and bust cycle?
Before we go any further, let me tell you about this week's video sponsor, TailorStore,
where I've been getting my shirts from for more than a year.
TailorStore make custom clothes designed for your body, so everything fits perfectly.
You enter a few details on their website, they calculate your size and make a custom
shirt for you.
If it's not right the first time, they'll remake it for free.
You can pick one of their designs or use their software to design your own shirt, specifying
collars, cuffs, fabrics, buttons, even monograms.
I'm wearing their phoenix white shirt with cutaway classic collar, convertible cuffs
and mother of pearl buttons.
At TailorStore, nothing is produced in advance, so there's no unsold stock and far less
waste.
And because they own their own production facilities, there are no middle layers pushing
prices up.
What you're paying for is quality.
If you want a smarter way to shop for clothes that actually fit, click the link in the description.
TailorStore is offering my viewers 25% off their first order with the link and discount
code in the description.
Custom clothes at a price that actually makes sense.
Okay, so let's say you were worried about AI investment and bought small cap US stocks
instead.
Well, you're doing quite well.
The Russell 2000 had its best first half since 1991, up around 22%.
You might take that as a sign that small, ordinary American businesses are booming.
In reality, 16 of the 50 best performers in the index are semiconductor and chip equipment
firms.
Companies like MaxLinear and Airtest Systems are up 250 and 380% respectively this year.
You didn't dodge the AI boom.
You bought the companies that sell it cables and testing gear.
Or maybe you're a sensible value investor.
You don't touch overpriced growth stocks.
You bought the Russell 1000 value index.
Right now, you feel pretty clever because value is up about 20% this year while the
growth index has actually fallen.
Value investing, it seems, is back.
The reason your value fund did so well this year is that until recently it was packed
with high-flying semiconductor stocks.
Micron, AMD, Western Digital, which had been on an enormous tear.
Then, in late June, the index providers did their annual rebalance.
They moved those chip stocks out of the value index and into the growth index and moved
Amazon, Apple and Microsoft the other way into value.
And the timing was, by complete accident, absolutely perfect.
The chips got sold right at the top of their run, just before they rolled over and the
value index picked up the big tech names right as they were touching their lows.
Nobody made any big decisions here.
A calendar reminder did it.
A value fund manager quoted in the Wall Street Journal could only stand back and admire how
it had worked.
It was like the index got so lucky, he said.
It caught that blow off top in momentum and then sold it right before they rolled over,
which I imagine is a slightly painful thing to say out loud if your actual job is picking
value stocks and a mechanical rebalance just did a better job than you.
So, as a value investor, you did well twice over.
You made money on the chips and sold them at the top.
The only small catch is where it left you.
Your sensible value portfolio is now stuffed with Amazon, Apple and Microsoft, three of
the biggest tech companies on earth and, quite possibly, the exact stocks you bought a value
fund to avoid owning.
You escaped the AI trade and landed in what was effectively the AI trade wearing a false
moustache.
So, if you're stuck in this trade, whether you want to be in it or not, the obvious question
to ask is how much money are we talking about?
If the AI bubble deflates, how much wealth is likely to be wiped out?
Well, let's start with the economist Dean Baker, who runs a website called the AI Bubble
Monitor, which sounds like a relaxing website to check over your morning coffee right next
to the weather report.
Baker's arithmetic goes like this.
The total US stock market is worth around $80 trillion.
If price-to-earnings ratios simply drifted back to their long-run average, not a crash,
just a return to normal valuations, that would erase something like $40 trillion of stock
market wealth, which he notes averages out to nearly $300,000 per US household.
Now, we need to be careful with the per household figure like that, as it's an average and averages
can be sneaky.
Stock ownership in America is wildly lopsided.
The richest 10% of households own something like 90% of the shares.
So, if Elon Musk lost say $200 billion from SpaceX and your neighbor lost $4000, the average
loss is a very impressive number that describes almost nobody.
The typical household would lose far less for the simple reason that the typical household
doesn't have $300,000 in stocks to lose in the first place.
So, let's get some other opinions.
Gita Gopnath, the former chief economist at the IMF reckons a dot-com style correction
today would destroy about $20 trillion of American wealth, plus another $15 trillion
in wealth held by foreigners.
That American $20 trillion is roughly 70% of US GDP.
Consultants at Oliver Wyman ran their own version and landed near $33 trillion of value
wiped out, again more than the entire US economy produces in a year.
And then, for a number that puts those estimates in perspective, when the actual dot-com bubble
burst, it destroyed about $6 trillion of equity value.
So, the mainstream estimates for this one run somewhere between 5 and 6 times the size
of the crash many of us still reach for as a cautionary tale.
So, three different methods that all land in the tens of trillions of dollars.
We can at least agree on the order of magnitude of the potential hole.
Now, you might reasonably say, so what?
It's paper wealth anyhow.
If your retirement account falls by $100,000, you've no less cash in your pocket this morning
than you did yesterday.
That is true, but it matters anyway, and it matters more now than it used to.
According to Goldman Sachs and the Federal Reserve, stocks overtook real estate as the
single biggest component of American household wealth this year, for the first time since
the Second World War.
For most of modern history, the average family's net worth was mostly their house.
So, a stock market crash was mainly a rich person's problem.
That is no longer the case.
The stock market is now where the median household's wealth actually lives.
Which means a crash today reaches into the middle of the country in a way that it didn't
in 2000.
That's then where the wealth effect comes in.
When people log into their brokerage accounts and see a big number, they feel rich and they
spend more.
The nicer car, the kitchen, the holiday.
The research rule of thumb is that for every $100 of paper stock wealth, people spend about
three real dollars in the actual economy.
It may not sound like much until you multiply it by tens of trillions of dollars.
And the whole thing runs in reverse too, just as reliably.
If 30 odd trillion dollars of household wealth were to evaporate, people would drastically
slash their spending.
The new kitchen gets cancelled.
The contractor loses his job.
The car salesman has a dreadful quarter.
The dealership lays someone off.
The damage cascades out from the stock market into the real economy, without a single bank
having to fail.
It just requires people to feel a bit poorer.
The problem for anyone who thinks they're safely on the sidelines is that AI is now
propping up whole chunks of the non-AI economy.
Harvard's Jason Furman calculated that AI-related infrastructure spending accounted for something
like 90% of America's economic growth in the first half of last year.
More conservative estimates put it at around a quarter.
Either way, an enormous share of recent economic growth is simply just data center construction.
So think about who depends on that spending continuing.
It isn't just Nvidia shareholders.
It's the electricians we talked about at the top of the video.
It's the construction crews pouring the concrete.
It's the firms that make the air conditioning units that stop the servers from cooking themselves.
It's the people running the cables, manufacturing the transformers and driving the trucks.
Even the astronauts installing them in outer space.
If AI capital spending slows, all of those people see their income slow with it.
And not one of them needs to have ever bought a single share of anything.
The noble laureate Joseph Stiglitz puts the grim vision plainly.
A crash like this would land at the same moment AI starts displacing workers.
So a lot of households could get hit twice at once,
their savings falling while their jobs get less secure.
As he put it, the breaking of any bubble is really bad in the short term for the macro economy,
which from a Nobel economist is practically a scream.
Now, if your mind is going to 2008 at this point,
you might be worrying a bit too much.
The big banks are in much better shape than they were back then.
They hold far more capital, they've got real liquidity buffers,
and they get stress tested every year.
There's no reason to expect a repeat of Lehman Brothers.
But the reason it isn't a banking story is that we made the bank safer.
And when you make one part of the financial system safer,
the risky lending doesn't necessarily stop.
It just moves somewhere else with fewer rules.
A few weeks ago, the Wall Street Journal published an analysis with the wonderful title
Why Big Tech's AI Spending is $3 Trillion Higher Than It Seems.
They reported that every quarter the big tech companies
proudly report their capital spending on AI, the data centers, the chips, all of it.
That reported figure across the whole group is about $600 billion over the last year.
Here's the problem, though.
The Journal went through the footnotes of these companies' filings
and found roughly $3 trillion of additional AI commitments
that don't appear on the balance sheet at all,
five times the capex everyone's been staring at.
These are things like long-term leases on data centers
and locked-in purchase commitments for chips,
computing power and energy,
money that the companies are absolutely on the hook to spend tucked away in the footnotes
rather than sitting on the balance sheet where you might think you would find it.
Alphabet alone has over $800 billion of this type of purchase commitment.
Analysts have started calling it an AI spending iceberg.
The enormous part above the water turns out to be the small part.
None of this is technically hidden.
And none of it is illegal.
It's all disclosed if you happen to be the sort of person who reads footnotes.
But it means that the true scale of what these companies have promised to spend
is far bigger than the headline numbers suggest.
And every dollar of it rests on the assumption
that the AI revenues will eventually turn up to pay for it.
The market has started to notice.
The cost of insuring big tech's debt against default has hit record highs recently.
And when the companies at the center of all this start to strain,
the pressure travels to wherever the lending happened,
which brings us to private credit.
Over the past few years, private credit funds,
lightly regulated outfits that lend directly to companies outside the traditional banking system,
have poured money into the AI and software world.
This is now something like a $2-3 trillion market.
And it has started to creak.
The Financial Times reported this month that troubled loans at the 20 largest
listed private credit funds have climbed to their highest level since 2017.
Fitch says private credit defaults hit a record in July.
One large fund reported that 7% of its entire loan book was in trouble.
The co-head of one of the big lenders told investors,
more or less, that the denial phase is over.
Now, we have to be fair.
Many people think that this is overblown too,
pointing out that the default rates we're talking about are still low in absolute terms
and if the lenders recover most of their money,
the actual losses are still small.
So there's no catastrophe today.
The worry, however, isn't the current number,
it's the direction the number is moving in
and the fact that private credit is deliberately hard to see into.
In a lightly regulated opaque corner of finance,
nobody quite knows who's holding the risky loans
until something goes wrong and everyone finds out at the same time.
We ran this exact experiment in 2008 with a different set of acronyms
and it didn't go especially well.
We're now rerunning the experiment,
presumably to confirm that the result replicates.
Now, whenever you point any of this stuff out,
someone always turns up in the comments within about 90 seconds
to say that none of this matters
because AI is a real technology that's going to change the world
and they may well be right.
There might be huge productivity gains in the pipeline.
Businesses, just to be clear, already find AI really useful.
But here's the uncomfortable thing.
The technology being real and useful and world-changing
still does absolutely nothing to protect you
if you initially overpaid for the stock.
Good businesses and good investments are totally different things.
The Bank for International Settlements put out a report recently
comparing the current AI build-out to the British railway mania of the 1840s
and they weren't being flattering.
Trains were obviously a real and transformative technology.
They changed the world, how people travelled,
how goods moved, where people lived.
Investors in the 1840s who understood how important this technology was
got extremely excited and did exactly what investors always do,
which is to take a good idea and overdo it.
Hundreds of new railway lines were proposed
and the money involved was staggering.
By 1850, cumulative investment in railways
had reached close to half of Britain's entire GDP.
The snag was that in all the excitement,
companies started laying down expensive railway track
to tiny villages that turned out to contain almost no passengers.
Amazing engineering, but nobody on the platform,
which as a business model has some well-documented weaknesses.
By 1850, railway shares had lost about two-thirds of their value.
The tracks remained in place, the trains ran,
the technology went on to power a century of British industrial dominance,
but the people who paid for it were wiped out.
The railways changed the world,
the railway investors changed their spending habits.
This pattern is what the BIS is worried about.
According to their research, relative to where each boom started,
the AI build-out has already grown faster than the railway mania,
faster than the electrification boom of the 1920s,
and faster than the dot-com bubble.
On their chart, it's the steepest line of the lot.
Those earlier manias tended to break around year five
and then drag their economies into a recession.
We're currently at year three, so there's plenty of time left to go.
For a more recent example, we can look at the dot-com bubble.
The internet, once again, was real.
No one's claiming that it turned out to be a fad.
And at the time, telecom companies were so certain
that the future would need infinite bandwidth
that they borrowed huge sums of money
and laid tens of millions of miles of fiber optic cable across the country.
So much of it sat unused for years afterwards
that the industry gave it a name, dark fiber.
They built the road system for the modern internet
roughly a decade before there was enough traffic to justify it,
and most of them went bankrupt waiting for the eventual revenues.
When that bubble burst, the S&P 500 fell by about half
and the NASDAQ lost nearly 80%.
But here's the part that really matters for anyone who thinks
they can outsmart this by simply buying the eventual winner.
Even if you picked correctly,
even if you identified the single best company of the entire era,
it could still ruin your decade.
Amazon is the obvious example of a winner that you could have picked.
Amazon survived the crash and became one of the most valuable companies
in human history,
but its stock still fell about 90% when the bubble burst.
If you bought at the peak in 1999,
you were absolutely right about the future of online commerce
and you then would have had to wait until 2009
to get back to break even on your investment.
Being right cost you a decade.
Jeff Bezos's own letter to shareholders about that year
opened with a single word,
ouch.
It's worth noting that there are very few people
who would have bought just Amazon either.
In the typical tech investor's portfolio in 1999,
along with Amazon was a basket of stocks like pets.com
that all went to zero,
diluting their long-term returns significantly.
For investors, the hard part isn't believing in the technology.
It's picking which specific company is going to win
while you're standing inside the bubble.
If you cast your mind back to the late 90s
and you're convinced the internet is the future
and you're right about that
and you started looking for a company
that will own internet search,
the obvious choice at the time
might have been Infoseek or Lycos or AltaVista or Excite,
the biggest, best-funded search leaders
that had become household names.
Every one of them is now a trivia question.
The company that eventually won, Google,
barely existed during the bubble
and only came to prominence in the early 2000s
after the bubble had burst.
It's very easy looking backwards to assume
today's winners were always destined to win,
but history is fairly blunt on this topic.
Being first to build a transformative technology
mostly just makes you a very expensive rough draft
for whatever shows up later and actually makes the money.
Now, before you take any of this as a signal
to sell everything you own to buy canned food and guns,
I wanna give you the other warning too,
as calling the top on tech
is a game people have been losing for quite some time too.
Back in April, 2022,
the New York Times ran a piece called
The Tech Bubble That Never Burst.
It laid out a full decade of famous investors
sounding the alarm on tech
and being wrong over and over again.
In 2011, the entrepreneur Steve Blank announced
we were in a second internet bubble
and that the signals are loud and clear.
In 2014, the VC investor Mark Andreessen
warned that high burn startups would,
and he wrote this in all caps, vaporize.
In 2015, Mark Cuban declared
that this bubble was worse than the tech bubble of 2000.
In 2016, the veteran investor Jim Breyer
saw blood in the water,
predicting that 90% of unicorns
would be repriced or die.
And in 2021, Jeremy Grantham, a legendary investor
who has correctly called some of history's great bubbles,
promised that this one too would burst in due time.
The bit I love though,
is that the New York Times published that article
in April, 2022,
which was more or less the exact month
the tech stocks began
one of their worst declines in a generation.
The Nasdaq fell by about a third that year,
so the definitive piece on doom-mongers always being wrong
came out at almost the precise moment
the doom-mongers were briefly right.
And that is the whole problem in one story.
The bears are usually early, frequently wrong,
and then occasionally, with no warning, they're right.
And by that point, most people have stopped listening,
which is exactly why this video
isn't a prediction,
and why sell everything is not the lesson.
In fact, if you had bought at the highs in 1999,
or right before the credit crunch,
or right before COVID,
and just held on till now,
you will have done just fine.
And the problem with selling early,
or even selling in a timely manner,
is that it's very difficult to time rebuying.
So if nobody can time this,
the lesson is something much more boring than that.
Jason Zweig wrote a column in the Wall Street Journal
a little while back,
making the case that if you actually want to get
some distance from the AI trade,
one place to look is Europe.
He was not especially flattering about it.
He called the continent, and I'm quoting here,
an open-air museum of aging populations
and arthritic economies,
which is a rough sentence to read if you live in Europe,
and would make for a fairly grim tourism campaign.
But the unflattering description
is sort of the whole point.
European stocks are cheap
precisely because so few people are excited about them.
Tech is only about 10% of the European index
versus nearly half of the S&P 500.
And value investors are looking there
exactly because the excitement
and the high prices that come with it
simply hasn't arrived.
Instead of AI and Mars colonization,
the index is mostly banks, industrial manufacturers,
and healthcare companies,
the sort that generate steady cash
and pay you a dividend of around 3% while you wait.
And to be fair to Europe,
this isn't just buy the boring stuff and hope for the best.
Zweig quotes a fund manager at Fidelity
who describes the 3% dividend
as the ballast you'd look for
if the AI story fails to meet expectations.
The idea being that if the S&P 500 falls,
European stocks will probably fall less
simply because there's less speculative price buildup
to unwind, a port in the storm as he puts it.
Another manager Zweig spoke to argues
the ossified Europe stereotype is out of date,
that big European companies
are reforming and cutting costs faster
than she's seen in her entire career.
So the case isn't that these are sad, dusty companies,
it's that low investor expectations
leave room for upside surprises,
which is roughly the opposite of what you get
when you buy the most exciting stock in the world
at 40 times earnings.
Now, I should be clear that Zweig
is not predicting a crash in his column either.
He goes out of his way to say
the fears about US market concentration
are probably overblown,
that history doesn't show
that a few giant stocks lead to bad returns,
and that if you're already globally diversified,
you might not need to do anything at all.
His suggestion is simply
to add a little European exposure at the margin,
not to run for the hills,
which fits the whole spirit of this video.
The point was never to tell you
that the sky is falling.
It's that if you've somehow ended up
with everything you own riding on a single trade,
it's worth knowing that there are
other rooms in the building.
And that brings us back to the magic word
from the start of the video, diversification.
People often think diversification means
owning 50 different technology stocks
that all go up together.
That isn't diversification,
that's just piling in in a bull market.
As Zweig puts it,
if all of your assets go up at the same time,
they're also likely to go down at the same time,
which is the thing you were trying to avoid.
Real diversification means deliberately owning things
that don't all fall over on the same afternoon,
which in practice means holding some assets
you feel faintly embarrassed about.
While your neighbor is at the barbecue
talking about his SpaceX allocation
and his Nvidia call options,
you have to admit that a meaningful chunk
of your net worth is riding on the dependable performance
of a Swiss pharmaceutical company
and a Scottish water utility.
It's not a story that gets you invited back,
but the boring assets are the ones still standing
when the exciting ones aren't.
And that's the real lesson here,
because long-term investing isn't supposed to be exciting.
When you look back through financial history,
the most thrilling trade of any given era
is very often the one that goes on
to ruin the people who piled into it.
In the early 1970s, the nifty 50
were the 50 blue-chip American stocks
that you could supposedly buy and hold forever,
right up until they lost most of their value
in the 1973 to 74 crash.
By the end of the 1980s,
the Japanese stock market was widely regarded
as the unstoppable future of the global economy.
Then it spent the next three decades
going essentially nowhere.
An investor who bought Japan at the peak in 1989
waited more than 30 years just to get back to break even.
And as we've already discussed,
the NASDAQ needed about 15 years
to climb back to where it stood in the year 2000.
None of these were stupid bets at the time.
They were the consensus.
They were the exciting, obvious,
everybody-already-knows-it trade,
which is precisely what made them so dangerous.
So, this is where we land.
You don't have to attend every party.
You don't have to put your whole net worth
into the hottest corner of the market
because it's what everyone is talking about
and someone on the internet told you it only goes up.
The AI boom may well work out to be as good as advertised.
A lot of serious people think it will,
and they might be right.
But this technology is real
and these stocks are a good investment at these prices
are two completely different statements.
And a good chunk of financial history
is just people confusing the first one for the second.
The point of investing for retirement
isn't to have interesting things to say at dinner parties.
It's to eventually retire.
If you found this video interesting,
you should watch this video
on what's been driving the rise
in long-term borrowing costs next.
Don't forget to check out our sponsor, Taylor Store,
using the link in the video description.
See you in the next video.
Bye.