Okay, so Sundar Pichai walks up to you.

You're walking the streets of Los

Angeles. You're on Skid Row. I don't

know how you end up there, but you you

take a wrong turn. You're walking down

Skidro. There's the 10 cities and Sundar

Pichai gets out of his tent, CEO of

Google, but he's, you know, falling on

rough times. He's disheveled. He's got

the college shirt on, but he's dirty.

He's disheveled. You know, he looks like

he hasn't eaten, hasn't slept, and comes

up to you and he's like, "Hey, man. Hey,

man. Can I borrow $15 billion, please? I

just need it. Uh, I just need it. Can I

borrow 15 billion, please? That's

essentially what's going on. This is

breaking news this morning where Reuters

just reported about an hour before I'm

sitting down to record this that Google

is doing a $15 billion bond sale, which

is basically them saying, "Hey, we need

some crowdfunding to power our AI data

center spending binge. This is not a

good look for things to come this year."

So, they're selling bonds that won't

mature until 2066. It's a 40-year bond,

which means they're asking investors to

take the bet that AI infrastructure will

still be providing a high return on

investment when most of the executives

at those companies will be long

deceased. The figures on that, it's a

$15 billion high-grade bond sale

announced today, just about an hour ago,

over seven tanches. Longest tunch

maturing, like I said, 40-year in 2066,

1.2 percentage points above treasuries.

This is their second bond sale in 3

months. They did one back in November

that was 17.5 billion and they also sold

6.5 billion to Europe and that's dollars

and euros respectively, whatever the

euro is. And if you haven't been

tracking this, this is part of a broader

theme going on in tech right now. Google

specifically basically they said, uh,

hey, we're looking at spending about 120

billion in capex. That was kind of the

Wall Street estimate leading into their

earnings last week. And then last week

during their earnings they said, "Hey,

actually it's going to be 60 billion

more. We're going to spend 185 billion.

60% of that is going to go to the

servers and GPUs and then 40% of that is

going to go to the hardware,

infrastructure, networking, those sorts

of things." You might be like, "Dr. J, I

like I don't know. I'm not like a big

baller executive. Like I've never been

at the adults poker table. Like what is

that even a lot of money? Like that just

sounds like a business amount of money."

Like I And yes, that's a lot of money.

Let's put it in relativistic terms here

because it is a lot of money. So again,

they're guiding around 175 to 185

billion in capex this year. That's the

capital expenditures. That's the money

they're going to pay on stuff like data

centers, primarily data centers. That's

double what they spent in 2025, which

was around 90 billion, and it's triple

what they spent in 2024, which was only

52 billion. So even for Google, like

this is insane. Wall Street was not

happy with this either. When Google

said, "We're going to be looking to

spend around $185 on AI data centers

this year," Wall Street was like, "Are

you fist me right now?" And the stock

dropped 5%. It since recovered, but it

dropped 5% right after that

announcement. And it just put that 180

billion in context for you. Bespoke

Investment Group had this great quote.

They said, quote, "There are only 59

other S&P 500 companies that Alphabet

could not buy with 180 billion in

capex." So 59 companies aside, every

other company on the S&P 500 listed on

the S&P 500 with that 180 billion, they

could just basically throw a dart at

that chart and just buy that company

with that amount of money. So even for

the class of companies and the size of

companies they're in, this is an insane

amount of money. And if you want the

clear statistics on this across all of

the companies where this AI spending is

going this year, get my free AI spending

tracker. The link is down there in the

description. You see that Google isn't

an outlier here. This is part of a

broader pattern that seems to be

accelerating over the past few weeks in

terms of basically crowdfunding issuing

these bonds. The five hyperscalers

issued $121 billion in bonds last year,

which if you average across what they

issued in bonds between 2020 and 2024,

that's about a 28 billion per year uh

bond issuance per year. So this is a 4x

increase in just one year for what's

typical for all five of these

hyperscaler companies. You have Oracle

raising 25 billion just last week. Meta

with 30 billion in October, which by the

way is the largest non-merggers and

acquisitions bond sale ever. And again,

if you've been following the channel,

you know that the combined hyperscaler

capex, the projections on what these

hyperscalers are going to spend on AI

infrastructure and data centers this

year is over 600 billion. The best

number we have on that right now is

around $660 billion. Even if we drill

down into the bank data, all of this

backs us up. Bank of America securities

data, 121 billion bond issuance in 2025

versus 28 billion a year average between

2020 and 2024. Morgan Stanley, JP Morgan

project 1.5 trillion in new tech debt

over the next few years. And this is

really the thing you need to keep in

mind. If all these buzzwords are

floating around, the one the one thing

that they think you're too dumb to

understand, and this is the core of

what's going on, is they are spending

more money than they are earning. Okay?

So, it's like if you have a lemonade

stand and you sell a few lemonades per

day and you walk away with 15 bucks from

your lemonade stand. It costs you $10 in

ingredients to make a batch of lemonade,

but you decide you're going to take a

loan from your mom and dad because

you're 5 years old and you have a

lemonade stand. And you're going

to do two batches. You're going to need

a $5 loan that you repay them with

hopefully the profit you make from the

second batch of lemonade. Although you

don't know for sure if the demand is

there in this case. So this is something

that helps public companies scale faster

when done responsibly. This is like a

this is a normal thing when you can say

hey if I make that second batch of

lemonade here's clear user demand.

Here's clear signal that we're going to

sell out of that lemonade and you can

pay that bond back. No problem. And you

have a long time to pay it back. As

we've discussed before we got a great

video on what's going on with Meta and

their private internet that was just

posted yesterday. This has happened

before. During the dotcom bubble, the

telecom industry invested $500 billion

in fiber optic cable. Those companies

built the fiber optic cable that was

supposed to wire us for the new

internet. Then they went bankrupt. Those

cables sat dark and unused and then they

were bought for pennies on the dollar by

other companies. The internet still won,

but investors got destroyed. This is not

tinfoil hat. The e com editor explicitly

drew the fiber optic parallel in their

journal. tech, equipment, and software

development has hit about 4.4% of our

GDP, which almost matches the.com peak.

And the numbers are really screwy on

this. I mean, it's got to be a big

return on investment. Nobody can prove

that yet. These AI assets depreciate at

about 20% a year. So, if we take these

tech companies at their word and

projected growth, that's $400 billion of

annual depreciation, which is more than

the profits. So they like in the

lemonade stand example, they don't even

have good evidence that this is going to

pan out or that they have adequate user

demand for that extra lemonade. Here's

what nobody in the techf funded media is

willing to tell you. Google and the

other hyperscalers are now taking

crowdfunding. They are selling bonds to

fund AI infrastructure development which

has not proven that it can generate

returns at scale. It is vibe investing

is what's going on. You get a sense that

the fear underlying all hyperscalers

right now is something akin to you can't

slow down or you'll lose. And I'll tell

you that is the underlying fear of every

bubble leading up to every bubble. Every

company thinks that it's an emotional

argument. It's not a business argument

based on stats. So this isn't about

whether or not AI is valuable. We can

continue to hash that out in the

comments as many of you have been doing.

It's its own video series on its own.

But the argument that I'm making here

and I I hope that you take away from

this is that the tech sponsored media

wants you to think that this is a sure

thing that AI is going to provide such a

massive return on investment that these

seemingly crazy financial decisions are

actually matter-of-act and well sourced

and they're not. This is being done on

Vibe. There are no numbers to back up

the claim that this will be as

successful as they say it is. Again,

it's not a debate about the raw

principles of AI. This is a debate about

how fast they're scaling and the poor

business and financial decisions they're

making along the way. Thank you so much

for watching. Don't forget to subscribe

and please share this video wherever

your tech friends hang out on X, on

Reddit. Every share helps the channel.

Thank you so much for watching and I'll

see you in the next