What if I told you that there's an AI

tech giant hiding in plain [music]

sight? A company that builds the most

critical components for the AI

revolution, and it's even more

profitable than Nvidia. It's also

growing faster and priced cheaper than

almost any other AI company,

>> [music]

>> making this a great stock to get rich

without getting lucky. Your time is

valuable, so let's get right into it.

You probably already know the name, and

you might even own the stock, but

there's a good chance that you're

thinking about this company the wrong

way. Just like most Wall Street

analysts, last week Micron reported the

best quarter in their 48-year history.

Revenue came in at $41.5 billion for the

quarter, which is up by 74%

quarter-over-quarter

and 346%

year-over-year. Let me say that again,

Micron's revenue almost doubled in the

last 90 days, and it more than

quadrupled over the last year. And their

revenue guidance is $50 billion, which

would imply revenue growth of 342%

year-over-year for next quarter as well.

Gross margins came in at 84.9%.

That's not just higher than Nvidia's,

it's higher than most software companies

like Meta Platforms. In fact, it's the

highest gross margin that any physical

chipmaker has ever publicly reported.

When it comes to gross margins, Micron

is the current king. I've been covering

Micron for years now, and it's my second

best-performing stock of all time, only

behind Nvidia. Here's what I think most

Wall Street analysts have been missing

this entire time. The margins and the

revenues are not the story here. They're

the proof that memory is no longer

cyclical, and it's no longer a

commodity. On this latest earnings call,

Micron announced that they signed 16

strategic agreements with their largest

customers, all of which run through

2030. These deals are structured so that

customers either buy the chips that they

committed to at the price they agreed,

or they pay for the chips anyway. And to

secure their spot in line, those

customers paid Micron $18 billion

before a single chip was shipped. So,

companies aren't just paying Micron for

memory chips, they're paying to reserve

the right to buy those chips in the

first place years in advance. This is

why it's so important to understand a

company's products, not just their

profits. These contracts carry a

combined minimum of roughly $100 billion

in revenue at margins that are much

higher than at any point in Micron's

history. This is what I think the market

hasn't priced in yet. Over the next 5

years, Micron is contractually

guaranteed to earn their best margins

ever. But before we can dive into these

strategic contracts, we need to

understand why the biggest AI companies

on Earth are paying billions of dollars

just to reserve Micron's memory in the

first place. The science behind this

stock. So, let me show you why Micron is

still one of the most undervalued AI

companies on the market without all the

industry jargon. The Nvidia H100 is the

chip that trained most of the AI models

that we use today. It sits on a circuit

board that's roughly the size of a

textbook. It uses 700 watts of power,

which is about seven light bulbs worth.

And it can get so hot that the cooling

hardware can weigh more than the chip

itself. For years, Wall Street analysts

focused on the GPU, the chip that does

the number crunching and turned Nvidia

into the most valuable company on the

planet. But look at what surrounds it.

Arranged in a tight cluster around the

GPU are tiny towers of memory, each one

smaller than a fingertip, and each one

holding 16 GB of memory built in a

fundamentally different way than the

sticks of RAM that we cram into our home

computers. This is high-bandwidth

memory, or HBM, and it matters more than

almost anything else in the entire AI

buildout. Modern data centers don't

struggle to do the math behind AI. They

have thousands of interconnected GPUs

with thousands of compute cores each.

The real struggle is feeding those GPUs

data fast enough to keep them fully

utilized. An idle GPU isn't just wasting

time, it's burning money. Buying more

GPUs doesn't solve this problem. It just

makes the problem worse. So instead,

data centers need memory that can feed

every GPU as fast as those GPUs can

crunch the numbers. Which means the

memory needs to have very high

bandwidth. High bandwidth memory solves

this exact problem by stacking memory

vertically, eight to 16 chips high, kind

of like floors in a tiny skyscraper.

Instead of the data traveling along a

circuit board, it moves up and down this

tower of chips, and the path it takes is

16 times wider than the memory we put in

our PCs. If HBM is a tower of chips,

then the copper columns connecting them

are like high-speed elevators. Each

column is narrower than a single human

hair, and there are thousands of them

carrying data from each chip in the

tower in parallel. As a result, one H100

can move over 3 terabytes of data every

second. Or said in English, that's fast

enough to move around 600 full-length

movies between chips every single

second. At a high level, that's the

science. Now, let's get back to the

stock. High bandwidth memory isn't just

another component on Nvidia's newest

Blackwell chips. It's almost half the

total manufacturing cost. Memory's share

of the total manufacturing cost climbed

from about 20% of the Nvidia Ampere

A100s to roughly 45% of the Blackwell

B200s today. That's why I keep saying

that memory is no longer cyclical, and

it's not a commodity. It's one of the

most valuable parts of the entire AI

build-out, which is why Micron has

become one of the the valuable companies

on the entire market. But there's

something else on the market that you

need to know about, and that's your

private data. There are hundreds of

online data brokers making big money by

collecting and selling your personal

information. That's why I've been using

this video sponsor, DeleteMe, for over 2

years now, and I can't recommend them

enough. DeleteMe is a hands-free

subscription service that will remove

your personal information from those

online data brokers. They give you a

quarterly privacy report showing

everything they've done, and they

reviewed almost 60,000 listings for me

so far. But, what really surprised me is

that these data brokers had way more

than just my private data. They had my

wife's and my entire family's, too.

That's another reason I really like

DeleteMe. They have a family plan, so we

can all have more control over our

personal data. So, if you care about

your data and your family's privacy, you

can get 20% off any consumer plan with

my code Symbol20 by going to

joindeleteme.com

/symbol20,

or with my link in the description. And

a big thank you to DeleteMe and to you

for supporting the channel. All right.

Like I said at the start of the video,

Micron just reported their best quarter

ever. Revenues are up 346%

year-over-year. Non-GAAP gross margins

came in at 85% and they guided to $50

billion in revenue for next quarter.

Under normal circumstances, a report

like that would be the perfect time to

sell this kind of stock if memory was

still cyclical. Then, this would

definitely be near the top of the cycle

for sure. For most of Micron's history,

they had close to zero contracted

revenue. Every chip was sold based on

the monthly or quarterly spot price,

just like any other commodity. They had

no customer commitments, they never got

paid up front, and sales were never

guaranteed. If Micron's forecasts for

memory demand were wrong, then they paid

a hefty price. Either they made more

chips than they could sell, or they

wouldn't make enough and drive down the

returns of all the infrastructure they

invested in. And forecasts are almost

always wrong. But Micron just flipped

the entire script. They just announced

16 strategic customer agreements that

run through 2030. These are not ordinary

supply deals. They're structured as take

or pay, meaning customers either buy the

chips that they committed to, or they

pay for those chips anyway. There's no

cancel button. These [clears throat]

aren't small orders, either. The

companies signing these contracts are

the hyperscalers, companies like the

ones building data centers that run

ChatGPT, Google Gemini, Microsoft

Copilot, and Meta's AI infrastructure.

They're locking in their memory supply

years in advance because they can't

afford not to have it, especially when

their competition does. 14 of those

contracts carry a combined revenue

minimum of roughly $100 billion,

and to lock in their place in line,

those customers paid Micron a combined

$18 billion up front. That cash is

collateral. If a customer cancels,

Micron keeps the money. But this isn't

just Micron taking advantage of

desperate customers. These contracts are

a win-win situation. Micron gets a price

floor, guaranteed revenues regardless of

how the market shifts over time. But on

the flip side, their customers get a

price ceiling. They won't have to pay

insane prices if the memory shortage

keeps getting worse and supplies get

even more constrained. That's what makes

these contracts a solid long-term

business model. Both sides have a reason

to sign. I also want to emphasize that

the $100 billion

is the floor price. It's the worst price

that Micron will be paid, no matter what

happens in the market. That means more

than just guaranteed revenues. It means

guaranteed margins, margins well above

the peak quarterly margins from any of

their previous cycles. And the market

hasn't fully priced this in yet. For the

last two decades, analysts that were

covering Micron couldn't model next

year's revenue with any confidence. The

company had no backlog and spot prices

changed every quarter. So, financial

models went obsolete every few months.

That's why memory companies used to

trade at a steep discount to the rest of

the market. But, these contracts get rid

of that uncertainty for a third of

Micron's revenue, and a bigger

proportion as time goes on, while Wall

Street's models are still catching up.

Micron just went from zero contracted

revenue to a hundred billion dollar

backlog with eighteen billion dollars

already in the bank. Again, proving that

high bandwidth memory is not a cyclical

business. And now that you understand

the science behind this stock and why

this latest earnings call was so

important, let me tell you why I'm still

buying it, even though it's already a

ten bagger. And if you feel I've earned

it so far, consider hitting the like

button and subscribing to the channel.

That really helps me out and it lets me

know to make more deep dives like this.

Thanks. Now, let's talk about my

investment thesis for Micron. Micron

currently trades at roughly fourteen

times their projected FY26 earnings.

That number is twenty-two and a half for

Nvidia and thirty-one and a half for

Broadcom, meaning Micron trades at a

decent discount relative to other

trillion dollar chip companies. And

based on estimates for next year's

earnings growth, Micron's forward PE

ratio is just 5.5.

That number is over twenty for both

Nvidia and Broadcom, even though both

companies are still growing fast. Let me

say that again. Micron's earnings over

the next twelve months are growing so

much faster than their market cap that

their forward price to earnings ratio is

less than six, even though they're

posting 346%

revenue growth, 85% gross margins, and

have a hundred billion dollar backlog.

In my opinion, Micron is still so cheap

because the market is still modeling it

the old way. The cyclical chipmaker that

peaks and crashes based on commodity

pricing. That made sense for the last 20

years, but it doesn't make sense today.

And it definitely doesn't make sense in

a world where the DRAM market, which

includes high-bandwidth memory, is

expected to more than triple in size

over the next 7 years, which would be a

compound annual growth rate of 18.6%

almost 50% faster growth than the S&P

500. And Micron's leadership says that

the HBM market specifically will grow

more than twice as fast as that at

roughly 40% per year through at least

2028. Micron's HBM capacity is already

sold out through the end of 2027. And

they can still only fill about 60% of

what their biggest customers are asking

for. This is honestly one of the

strongest setups for long-term growth

that I've seen in years. A $100 billion

backlog. HBM supply sold out through

2027. Margin expansion from 39% to 85%

in just a few quarters. And the stock is

trading at under six times next year's

earnings. The biggest risk to Micron is

that all three major memory suppliers

are expanding capacity at the same time.

So, if AI spending does slow down,

supply could outpace demand. I

personally don't see that happening with

how much every hyperscaler is increasing

their CapEx budget year over year. But

even if it does, remember, Micron gets

their $100 billion

thanks to the take-or-pay structure of

their strategic contracts with $18

billion already in the bank. So, Micron

is positioned to get rich without

getting lucky. And so are Micron

shareholders. Let me point out one more

thing because it's too important to

leave out. Everything we've covered so

far is the base case. It doesn't account

for what could come next. The 16

strategic contracts cover about a third

of Micron's current production, but they

have four new fabs either under

construction today or breaking ground

right now. Every gigabyte of new

capacity that Micron brings online is

another contract waiting to be signed,

not just by existing customers, but by

others who already know the cost of

being left behind. And while Wall Street

analysts keep calling for AI spending to

slow down every single year,

hyperscalers keep spending more. They're

not cutting back. They're building data

centers faster than everyone predicted,

and every one of those data centers

needs more high-bandwidth memory.

There's also the demand side of the

story. Every major breakthrough of the

last 3 years has done the same thing.

It's made AI cheaper to run, which makes

more people use it for more use cases

more often, which drives total compute

demand even higher than before. So, the

more efficient AI gets, the better an

investment it becomes for those

providing it, which means memory demand

will keep accelerating with the next big

breakthrough and the one after that and

the one after that. None of that is in

today's models. None of that is in

Micron's current $100 billion backlog,

and none of that is baked into their

forward P/E ratio of just 5.5. I've been

covering Micron stock for years now, and

this is the best setup I've seen for

their long-term growth. That's why I'm

still dollar-cost averaging into the

stock before Wall Street finally catches

on. And if you want to see more stocks

that I'm buying to get rich without

getting lucky, check out this video

next. Either way, thanks for watching,

and until next time, this is ticker

symbol U. My name is Alex, reminding you

that the best investment you can make

is in you.