The AI Industry Is YOLOing to a Bubble… And That’s Why It Won’t Burst.

All right, back in New York. Uh, let's

get this going. Stock market continues.

Uh, I'm going to spend a lot of time.

It's the end of the year outlook papers are going out. I'm not going to spend time on the traditional outlook because everything is dependent on AI. So,

everything that you get from these videos is what you're not going to get from strategists and economists and people focus solely on the broader part.

This is all about artificial intelligence and I think everyone knows that or else there wouldn't be so much bubble fears.

Uh I'm going to go through MPUs. I'm

going to go through the cognitive era to the kinetic era. Why this is so important for PMIs, what names are already showing it, how transports are a leading indicator for it. There were a

lot of videos. I'll go through some of the uh the best podcasts of the week. Uh

I'm going to talk about vibe coding. I'm

going to write a paper on it for 22V next week.

It's basically the chat GPT moment for software which has implications when you combine it with the NPU acceleration or

the VLA VLM acceleration. Uh Dario Modi gave a great interview I think for people who are both positive and negative. It's a must listen to uh

negative. It's a must listen to uh especially since the interviewer was uh Sorcin who just did a crash thing uh using AI as a reason for why a 1929

style crash uh is in the making uh or possibly. And uh at the very end I will

possibly. And uh at the very end I will go through the website launch that's going on. Uh I do want to say right now

going on. Uh I do want to say right now at the beginning I think this is very important as I'm going to really spend the time showing people uh how AI will help them both on the income side of

their uh life which is really their job but also on the balance sheet side. I'm

going to focus on the balance sheet uh balance sheet side to basically show people in in videos but also in the work that I'll be doing on a regular basis how that they can think about finding

names taking research uh from podcast and drill it down in the same way that I uh do this uh and Bitcoin is building a bottom right now. It'll be a big story

for next year. Let's get started. Uh,

after the fears of the AI bubble where we kind of were grinding lower and then we rip right back up in a very short amount of time back to the all-time

highs. Uh, again, I'll say it repeatedly

highs. Uh, again, I'll say it repeatedly this year. Every time there are fears

this year. Every time there are fears over AI as a bubble, you have to be looking for names at that point and focusing on the areas that work. Don't

listen to CDS stuff. Don't listen to debt stuff. that will start to matter in

debt stuff. that will start to matter in two years, in 3 years for sure as the revenue either does or doesn't come in.

But for the next year or two or three, you're going to be focused on the parts that I'm going to be going through today. Uh and it's better to start

today. Uh and it's better to start focusing on them now than waiting uh the way a lot of people did with chat GPT.

So IWM small caps going to be a major story in my opinion over the next four years because of the PMI made new all-time highs this week. Midcap right

at all-time highs. equal weight S&P also right at all-time highs. So this is not a cap weighted move. In fact, we still have Nvidia sitting down significantly from the highs and banks which I

highlighted as a tell that as long as banks are not falling down when people are worried about credit fears are worried about bubbles are worried about all these different things as long as banks are sitting near the all-time

highs and in particular Goldman Sachs and JP Morgan uh everything is fine. New

all-time highs for the banks, transports. Just look at these

transports. Just look at these candlesticks. Look at how many days in a

candlesticks. Look at how many days in a row now we've been closing above the opening price. And this really started

opening price. And this really started back there. This is a PMI tell. This is

back there. This is a PMI tell. This is

what you want to be focused on.

Transports are starting to show what commodities have been showing, which is this beautiful rounding longterm formation. Uh I've highlighted John Rog

formation. Uh I've highlighted John Rog on this. I'm in the same camp. You got

on this. I'm in the same camp. You got

the 200 day pointed up. You got stock price or commodities going higher. Uh

and it's not just gold. Uh you've got a lot of commodities that are going higher. And we've now hit the highest

higher. And we've now hit the highest level since early 23.

Junk spreads uh new all-time lows this week. So the tights are are continue and

week. So the tights are are continue and bonval had a little move higher and right back down. So we have nothing showing up in

down. So we have nothing showing up in fears over what's going to happen with rates despite people now calling again for shorting bonds. Uh Fed funds were

now at 95% of a cut for December. So we

had gotten down towards 30. We're back

up towards 95 and we have another one built in by the time we get into the April June period. At this point, uh there was a breath thrust, an almost breath thrust, but they use this on the

breath thrust. I only bring this up

breath thrust. I only bring this up because that's how strong the rally was.

And whenever you get these oversold conditions, this is uh what you want to look for. But also this just highlights

look for. But also this just highlights the ramping that is coming for a lot of names. When you have a breath thrust

names. When you have a breath thrust especially after everyone's complaining about breath uh which people were when it comes ripping out of here and you get the IWM making new all-time highs while

the S&P is just below. One year later breath thrusts are usually a great signal for forward gains particularly in participation. Now payroll data we did

participation. Now payroll data we did get the ADP. this is becoming more important despite what you may be focused on. And I say that because the

focused on. And I say that because the Fed has acknowledged it, but also because we've had so many revisions to the uh non-farm. Uh you can see the weakness that's been here now for six,

seven months. The headline out of this

seven months. The headline out of this or the I guess the guts of it. We had a negative number. We still have education

negative number. We still have education and health services along with leisure and hospitality. This is the baby boomer

and hospitality. This is the baby boomer side is up 46. So when you put them all together, it was a bigger negative number and it was driven by small businesses. We still have a K-shaped

businesses. We still have a K-shaped economy. We still have a problem for

economy. We still have a problem for small mom and pop businesses which are closing down. And we saw 120,000. Think

closing down. And we saw 120,000. Think

about that number. 120,000. It was the sixth of the last seven months that we've had negative job creation in small businesses. So anyone thinking that that

businesses. So anyone thinking that that rates are going to move higher, that the Fed is going to turn around, we have a problem in the economy. And the Kshape is not getting any better at this point.

And I do not see small businesses as being able to change things this quickly. This is a recession for a good

quickly. This is a recession for a good portion of the economy. Rates are too high for them. And AI is a direct competition. So it is very very dramatic

competition. So it is very very dramatic in terms of this and I think it's going to happen. I think you're going to have

to happen. I think you're going to have a move away from uh you know the broader consumption trends and it's going to be more isolated to the higher end. Uh, and

what you're left with is this continuing K shape. So, here's the job breakdown.

K shape. So, here's the job breakdown.

101 to 19 employees, 20 to 49, that's where all the losses. As you get into the midsize companies, they're still creating jobs. So, this is clearly an

creating jobs. So, this is clearly an issue. And here's the last seven months

issue. And here's the last seven months just staggering that we've had job losses continuously in small businesses.

This is a recession for small businesses. Now again, if you're focused

businesses. Now again, if you're focused on rates going higher, makes no sense to me uh to be, you know, you can sit there and say rates aren't going to go down.

You can sit there and say I want to play a steepening yield curve. But the

reality is tenure rates are right here.

And this is the pressure that's coming from the jobs market. These are all different indicators. This is not use

different indicators. This is not use the the ADP. This is the unemployment rate which is now to the highest uh level since the lows back in 23. And on

the back of that, we also have the inflation data. Here is the the final

inflation data. Here is the the final six months uh I'm sorry the the the final months leading into the tariffs.

Then in March when we started getting the liberation fallout we've had PCE core basically sitting down here continuously just take an average I guess of 0.2. So figure you're talking

about 2.4 to 2.5 in terms of uh an annualized core CPI. Definitely not

something to focus on in terms of 10ear yields which are up at four and change.

So wouldn't worry about that. Hits

coming in. I'm only putting this in there. So you guys can all go through

there. So you guys can all go through the the the things you think about it.

Uh you're going to have a lot of chaos in the Fed next year, but I think the the reality is it's very hard for people to not be airing on the side of cutting rates when we have a K-shaped economy

and the majority of people are suffering in terms of small businesses. This is

the key. So I wrote this piece this week. We are shifting from LLMs to VAS,

week. We are shifting from LLMs to VAS, from bits to atoms, from textual things to the physical side. Things that could

be made in seconds like this video, things that take time to get from the ground to make to build. We are entering a period where this infrastructure is being built as we speak. You want a

visual on it? We are leaving the cognitive era. We just spent the last

cognitive era. We just spent the last three years since ChatGBT rolling out massive data centers. We're not even in through the very early stages. We've

seen the capex numbers explode this year. This is those capex numbers are

year. This is those capex numbers are for this. This is the brain of the

for this. This is the brain of the cloud. We cannot go into the kinetic

cloud. We cannot go into the kinetic edge era which is where we're going now to put a brain in all of these products.

The brain itself is a PMI side. So, so

far we've just been teaching it in the cloud. That's why when you go on your

cloud. That's why when you go on your phone, you're using an app. This is just continuing what was happening from 2007 with the iPhone launch until now. Now,

what we'll be going through is shoving artificial intelligence in everything that we have.

This is the amount of investment roadmap and industries to focus on that were not part of this side. This is all just starting now. And I'm going to go

starting now. And I'm going to go through photonics. I'm gonna go through

through photonics. I'm gonna go through a bunch of different components. We've

seen the semis, but it's been really isolated to Nvidia. You see ASML is mentioned here. Pterodine will be

mentioned here. Pterodine will be mentioned in here. All kinds of names that I've talked about over the last three months that are having major rips, but there are PMI related companies that have already started. I'll go through

those. By the way, all of these visuals

those. By the way, all of these visuals were created in Nano Banana. If you

haven't used Nano Banana or you're scared of it or whatever you're not doing with it or you think AI doesn't do anything, if you have kids having trouble in school, take some text pages and just put in to Nano Banana and say,

"Hey, will you make this easier for my kids to understand?" Start using AI.

This is why I'll be launching a website to help people do this. This took about 10 seconds to build. Um, this is just a visual to make it easier to to see why the cognitive error of thinking high

latency training cloud center data compute constrained. This is all still

compute constrained. This is all still happening. I'm been using Google now

happening. I'm been using Google now about 70% of the time for a month. Now

all of a sudden they've hit compute constraint. I've been restricted

constraint. I've been restricted multiple times this week. This never

happened to me before in Google Gemini was happening to me on chatbt. So now

they've hit compute constraint because everyone's moving to Google Gemini. As

we talked about last week, the shift into bits and atoms again, robots, cars, phones computers appliances medical devices, everything will need these

things. And that means you're going to

things. And that means you're going to need the brains, you're going to need the nerves, which is where the photonics comes in. You're going to need the eyes.

comes in. You're going to need the eyes.

You're going to need the body in terms of the muscles. All of this stuff is coming and you still have to build out the data center infrastructure. So

people have been focused on this and wondering why the PMIs aren't going higher yet. They will start to focus on

higher yet. They will start to focus on them as they go forward. LLMs versus VLM versus VLA just so people get the amount of

compute greater than what we just are leaving now. So two to 10 times on the

leaving now. So two to 10 times on the VLMs, 10 to 100 as much as a thousand times for the VA's. So compute was dominant back there and flops. This is

where Nvidia dominated. Then we started getting into VLMs. This is when Micron SKH highinex and Samsung blew out. Now

you're getting into the latency in physics. This is why when you hear the

physics. This is why when you hear the bottleneck is energy. It's not just that. It's anything physical. The

that. It's anything physical. The

physical part of AI. These companies are tiny. They have been in a bare market

tiny. They have been in a bare market now for 17 to 20 years since China peaked and we had the great financial crisis. There's a lot of names in there

crisis. There's a lot of names in there that have just not benefited. Some of

them are showing up on here. Um, someone

reached out this week about Amberella in terms of, hey, is this a name that shows up on all of these names as you're doing your work and you're going through the sensor, a lot of these either they haven't moved or they're just starting

to move. But I highly recommend focusing

to move. But I highly recommend focusing on the data scarcity, the latency and power, the bandwidth, the actuation, energy. These are all indicative names

energy. These are all indicative names of that. If you want more specific

of that. If you want more specific things, if you want uh more in-depth research, call up 22V. Uh if you're a FA or a retail person and you want to get

more in the know of where this is, the website launch will hopefully be happening in January, you'll have access at a uh non-institutional level and I will have a video series that will also be on there which you can sign up for

where I'm going to teach people how to do this and keep adding to them as we go across. Now the playbook, I released

across. Now the playbook, I released this on October 7th. This was six weeks ago. Why NPUs are AI championship

ago. Why NPUs are AI championship moment. Uh this was using basically

moment. Uh this was using basically saying that MPUs are like audles at uh for a quarterback at the line. The

ability to think, see the defense, and then change what's going on. That takes

a lot more memory. That takes a lot more components. And that's why NPUs are

components. And that's why NPUs are critical for PMIs. Now, an upgrade cycle for cars, phones, computers, for PMIs.

Historically, upgrade cycles in cars, phones, and computers are major drivers of PMA PMI trend. So upgrade cycles where people are basically forced to buy something because the the gap between

what the current model is and the prior model is so dramatic that they're buying them. When it comes to phones and

them. When it comes to phones and computers, this is critical for your job. This is critical for success. But

job. This is critical for success. But

if you want to do I've had to purchase already this year two computers. I

purchased one last year. Before that I don't think I had purchased a single computer for about a decade. I needed

these because of the bandwidth side, because of the memory side, the demand side. All of these are going to happen

side. All of these are going to happen because AI and the brain are going to be on there. When you get an upgrade cycle,

on there. When you get an upgrade cycle, you get large order volumes, new orders come in, global supply chains are used, cyclical replacement platforms, in particular, autos are one of the strongest historical drivers of PMIs.

When people are thinking autos now, they're going, well, there are there are no auto sales. There's a robo taxi upgrade that's coming right now and that's going to lead to a roll out which requires massive sensors, compute, battery supply chains. If you haven't

gone to the Gigafactory or the car factories for Tesla, again, remember as these are rolled out in cities across the country, you're going to see massive needs for all of these components. This

is already starting in a small way, but we're at the very beginning stages of it. Why trucks will amplify the PMI

it. Why trucks will amplify the PMI effect. Autonomous trucks. Think about

effect. Autonomous trucks. Think about

every truck you see on the road. We will

be getting into autonomous trucks. It'll

start small, but it'll continue to go.

The reason I'm bringing this up now is robo truck rollout is a strong multi-year PMI boost. It's going to happen. It's going to happen for a lot

happen. It's going to happen for a lot of different components. You can't

ignore it and you can't be waiting saying, "Well, maybe it'll happen in 2027." The cars are going to go on. The

2027." The cars are going to go on. The

smartphones are coming. The computers

are coming. and you're already starting to see the build out on some of the component pieces which are also necessary for the data centers which are going on. So if you can find things with

going on. So if you can find things with a ven diagram where they benefit from the data centers but they're also going to benefit from the upgrade cycles of each of these the autos the phones and

the computers. Next year is the year

the computers. Next year is the year that this becomes a story. But it's also the year that people start to focus on humanoids. You want the picks and

humanoids. You want the picks and shovels for the humanoids before you get into the humanoids themselves. I expect

this to be a major theme. The impact of the robotoxy roll out what you're going to see and then eventually in 7 to 10 years it'll become a negative because we

won't need cars but that'll take a long time. How MPUs will drive the PMI

time. How MPUs will drive the PMI higher. Again,

higher. Again, MPUs will lift PMIs meaningfully over the next several years because the trigger a synchronized replacement cycle. And this is something that's

cycle. And this is something that's really important. A synchronized cycle.

really important. A synchronized cycle.

We've had a phone cycle. We had that with the smartphones. We had a PC cycle which came kind of after the smartphones. We've had uh a replacement

smartphones. We've had uh a replacement cycle for autos before. But the main part is that you're putting brains in everything. So every single thing, edge

everything. So every single thing, edge devices appliances enterprise hardware, industrial equipment, robotics with inside factories, it's going to be synchronized. You're going to have all

synchronized. You're going to have all these going on. None of these ever existed. So when you're looking for

existed. So when you're looking for traditional PMI cycles, I'm not seeing housing go up. I'm not seeing commercial real estate going higher. I see a bare market in this. Yes, this is why you have to think differently, you have to

remember that all of these things here are going to have brains in them. And

this is coming now. NPUs create a positive demand shock across multiple industries. Again, it extends across

industries. Again, it extends across industries. Everyone's going to be

industries. Everyone's going to be ordering more of these because it's going to make it easier and it's cost-effective. The amount saved for

cost-effective. The amount saved for having an autonomous truck over a driver and being able to do this stuff is going to be a major theme and you can't wait until it is here. We learned that with

Nvidia. By the time you started to

Nvidia. By the time you started to realize the buildout and the data center buildout for Nvidia, the stock was already up significantly and it didn't let people in. I'm going to show you some names like that now. Why this is a

2025 to 2029 theme. You guys can read this on your own, but basically VLM to VLA shift massively increases sensor and optical demand.

What does Gartner say about this? PC

shipments accelerating AIPC in 2025 2027. Smartphones moving to majority

2027. Smartphones moving to majority majority share by 2027.

Okay, remember if that's the case, PMIs will be up significantly over the next three years. Smartphones enormously

three years. Smartphones enormously influential during the 2010 2020 decade.

Again, largest electrics category in the world. It already had an impact in the

world. It already had an impact in the prior time coming out of the smartphones. And if you want to go read

smartphones. And if you want to go read about it, here's an IMF blog from 2018.

Smartphones drive new global tech cycle, but demand is peaking. You can go read this and go see on how much impact it had. In 2016, global smartphone sales

had. In 2016, global smartphone sales reached almost 1.5 billion units. They

were zero in 2006.

Corning. Here's one of those names that is benefiting from the data center side and will be benefiting going forward. So

Corning is a glass company. It's an

industrial loser. Software won. What

didn't win was physical stuff like glass. Once we got through the bulk of

glass. Once we got through the bulk of people getting smartphones, which again, you heard the peaking story there in 2018, basically the highs in 2006, just before the great financial crisis. Then

the iPhone gets launched right here went sideways unchanged from 2006 to 2023.

Then we basically double then we double again.

This company was created in 1851. This

is not an AI company but it is moving like Nvidia. This is not a company

like Nvidia. This is not a company created in the last 15 years. How does a company survive from 1851? Well, that

means there's not a lot of glass companies and that means there's not a lot of competition.

Why was their stock price unchanged?

Corning underperformed for 17 years because its biggest businesses were tied to low growth, highly cyclical, margin compressed markets, TV displays, telecon fiber, smartphones, while the company never captured the exponential upside

cycles that rewarded software and semiconductor peers.

If you need a visual for this, the stagnant years, their stock price goes down. None of these things are basically

down. None of these things are basically being used.

We've laid the fiber way before then everything else is done and we had a software where they didn't com weren't involved. Act two, their stock price is

involved. Act two, their stock price is rising and the reason it's rising is because of the data centers. They're

part of the optical growth needs in optical fibers that go into the data centers. What they're going to get as a

centers. What they're going to get as a future catalyst, the first one is the upgrade cycle for these. So the upgrade cycle for all of these will continue to mean glass sensors connectivity across

all three. And then you get this new

all three. And then you get this new massive buildout which is going to happen in humanoids. And again humanoids need lots of glass.

How many glass competitors are going to be able to serve this side.

Think about Corning stock is similar to Nvidia just at a different time period.

Here are their earnings. This is what they're talking about right now. Optical

communications. This was their springboard plan to get out. For them to grow, they need all those other businesses that are not moving to be smaller or less of a drag while this

grows. Reiterated in the last quarter,

grows. Reiterated in the last quarter, strong optical communication growth. So,

the data center side is driving their business, but they're going to have multiple exponential AIdriven capex cycles as you go forward. So, it was stuck for 17 years. Now, for the next six, it's being driven. And I say the

next six, it's only been two years so far. There's a massive story in terms of

far. There's a massive story in terms of what's going on in the upgrade cycle. It

would benefit significantly if uh if there was a synchronized upgrade cycle.

You'd have shortages. You'd have all the things you need. Pricing power to come back in. Corning is one of the few firms

back in. Corning is one of the few firms positioned across all three of these components.

How big a synchronized cycle could be when it's firing. GLW becomes a compound beneficiary. Think Qualcomm during 4G.

beneficiary. Think Qualcomm during 4G.

Broadcom during the hyperscaler billow and Nvidia during the early LLM phase.

More Corning stuff. The vision systems are where they're benefiting. They're

also benefiting from the optical fiber, the Gorilla Glass for the robot screens, which will be needed. Humanoids have

chest displays, wrist displays, and structural materials. I can go on and

structural materials. I can go on and on. You get the point. Marll, another

on. You get the point. Marll, another

company left for dead basically during the LLM boom, not really participating in a major way. Definitely going up and getting some benefit, but not Nvidia or

Broadcom style. And really since 2021,

Broadcom style. And really since 2021, during this period where PMI stayed under 50, they finally are breaking out and they're breaking out because they're finally starting to become part of the

data center build. And there's a variety of reasons. I'm not going to go through

of reasons. I'm not going to go through them, but the AI silicon, think photonix at this point. you're starting to get a different part of the semiconductors in there. I've been talking about it all

there. I've been talking about it all year. That's why I've mentioned

year. That's why I've mentioned companies like ASML, companies like Pterodine, companies that are a little bit more sensitive to the breadth of a uh semiconductor needs, which would be

again the PMI related stuff. That's why

those names are rallying and that's why the prices of DRAM are going up so much because the memory needs and the data management, the data density is changing. Celestial which is an

changing. Celestial which is an acquisition helps in the photonic side for Marvel and again generative clusters are increasingly bottlenecked by data

movement not flops this is the change from just being GPUs and needing us to move into a different stage that's why I've talked about the move from LLM what you need for this VLM which is where we

are right now and this is why we have a massive buildout going on which is why this is not a bubble we can't do any of this stuff until we have those data centers built or we can't do enough of

it. We can't get to this. Again, this

it. We can't get to this. Again, this

line, the VA, this is critical for the Genesis mission. This is critical for

Genesis mission. This is critical for the military. Both of these here,

the military. Both of these here, there's no way the government is going to let us get stopped here with a bubble. They will make sure we make it

bubble. They will make sure we make it to this stage and this stage. This is

not a guess. This is a reality. The

Genesis mission, this is what it needs it for. It's not

just the military side. It's for the biology side. It's for getting rid of

biology side. It's for getting rid of the deficit that we have, which I'll get through. We need breakthroughs in

through. We need breakthroughs in biology and chemistry and climate change, material science, and energy to allow us to get to the VA's. We don't

have that capability right now, which is why we need to spend the money on it.

Now, I believe factors are leading indicator on the PMIs. They start to tell you when the turn is coming. And as

I said last weekend, this is Momo, the Morgan Stanley momentum uh not sector neutralize. This is critical because it

neutralize. This is critical because it should be a sector related movement.

Here are the peaks. This down here is the new orders component for the PMIs inverted. When these have peaked and we

inverted. When these have peaked and we peaked here and we're now moving lower, this has been a turn in new orders for PMIs over time.

This is growth verse value. This gets

critical for mutual funds. UPM mutual

fund managers that are sitting out there, people managing growth funds, uh growth will continue to be a dominant factor. But I think the shift that's

factor. But I think the shift that's happening is happening so fast right now where companies which used to be say value uh or old school economy companies built you know founded in the 1850s that

hadn't done anything for a long time they quickly become growth names. How

quickly can the shift be made? The

market cap is really small in companies like Corning. It's a $70 billion company

like Corning. It's a $70 billion company now but that's after more than a uh a doubling or a tripling over the last few years. So it was a midcap company not

years. So it was a midcap company not that long ago. this growth versus value.

So far, it looks like we've peaked. It's

been a frustrating year for anyone on the growth side. I highly recommend mutual fund managers to come spend time with me on this growth side. I've talked

about this relentlessly uh with other people. I really do believe we're at an

people. I really do believe we're at an inflection point and I think the market is showing it. Transports are a leading indicator as I mentioned at the beginning for the PMIs. Their overall

chart looks great. These are the lines that I just took from the prior times of momentum and growth versus value, but in particular growth versus value. These

points, and this is the year-over-year transports. We just turned positive in

transports. We just turned positive in year-over-year transports. We just are

year-over-year transports. We just are higher than we were a year ago. John Ro

does great work on this stuff. I'll show

you a chart in a second, but the inflection point for transports occurs at the same point as growth verse value.

If we're going to have a shift in growth, transports are going to be a leading area. And here's the big base.

leading area. And here's the big base.

John Ro, one of his big bases. We

finally got the breakout. And again, we just finally took out one-year highs. We

were starting to do it at the beginning of last year. This break when PMIs were going up to 51, which is what they did here. Then we had the tariffs, we came

here. Then we had the tariffs, we came down. We're about to go higher, guys.

down. We're about to go higher, guys.

Here's the transport chart yearly candlesticks. I'm just highlighting

candlesticks. I'm just highlighting these because we did the same pattern this year. Took out the prior year's low

this year. Took out the prior year's low and then ripped higher. Took out the prior year's low, ripped higher. These

are all the same points that I basically mentioned. Took out the prior two years

mentioned. Took out the prior two years lows, ripped higher. Took out the prior year's lows, ripped higher. Did it again here. This is the way transports go.

here. This is the way transports go.

When they turn, it's on a dime. And

again, PMIs haven't turned yet. But

remember, there's a lot of really smart, big, quantitative hedge funds that pay for a lot of satellite imagery on the transports. When the transports start

transports. When the transports start moving and the trucks start going on the road, you ain't going to see the data before they are. uh SMH versus Nvidia.

Again, the breath of the market. This is

just showing you that SMH is back to the highs while Nvidia is still near the lows. We don't need Nvidia anymore

lows. We don't need Nvidia anymore during this PMI side. I wrote a Substack this week on Alon Musk versus Warren Buffett.

Old school, new school, trusted, not trusted. Both really wealthy. You got to

trusted. Both really wealthy. You got to start trusting this guy for those people have a bias on him. The most expensive bias you're ever going to have going

forward. Here's the interview. It's two

forward. Here's the interview. It's two

hours. Fantastic. Um, this gentleman got very philosophical with Alain in a Tesla plant. Uh, he Alon Musk can talk about

plant. Uh, he Alon Musk can talk about anything. I highly recommend regardless

anything. I highly recommend regardless of your view on Alon Mus to listen to it both as an investor but also as a parent about what the future's going to hold. I

want to highlight the 10 most provocative Alon Mus predictions that he said in here. Working will be optional within 10 10 to 20 years. The reason I'm doing the videos, the reason you need to

get your kids uh signed up, get them involved, college especially, working will be optional. We are not replacing all jobs with AI. But if people want to make the transition to where they have

the choice to be optional, there's two things they need. One is they got to focus on the income side of their life, which means their job. To be effective at work going forward, you need to be

able to use AI like a master. And to do that, you have to use it every day, which is why I'm going to focus my attention. At the same time, you have to

attention. At the same time, you have to focus on your balance sheet. You have to be making investments. You have to understand the markets. That is the second part that I'll be going through.

If you learn how to invest and you learn the markets, it will help you with AI.

It is the greatest resource for deep research for doing exactly what I do on these videos. Money will disappear as a

these videos. Money will disappear as a concept. AIdriven deflation will arrive

concept. AIdriven deflation will arrive within three years. Solarp powered AI satellites will harness the sun's energy. This changes the whole energy

energy. This changes the whole energy situation. And I highlight Google

situation. And I highlight Google announced a similar thing this week. I'm

bringing all of this up because this is what you need to focus on. AI and

robotics are the only path to solving the debt crisis. If you don't believe what Elon Musk is doing, if you don't believe what's happening, you are going to miss the VLM and the VLA side. He is

the leader in this, he has been preparing for 20 years for this moment.

20 years. He's been focused not on Tesla as a car company, which everyone calls it. He's been focused on it on something

it. He's been focused on it on something that only Adam Jonas talks about that has covered it, which is the ability of having the intersection of space, of

communication, of autonomy, of robotics.

Everything is in there and everything he's focused on depends on VLMs and VAS.

That is how you make money. So

unfortunately for those of you who hate Alon Musk, you're going to have to do your homework and focus on the fact that he is the information you need to listen to regularly. You can listen to

to regularly. You can listen to Moonshots. They cover a lot of the

Moonshots. They cover a lot of the things that Alon Musk covered as well.

Every time that I hear people talk about AI bubbles and I listen to these four guys speak MIT MIT I don't know where Naveen is from and I don't know where Sem is from but regardless go through

and listen to this on a weekly basis so you can just keep up to speed on what's going on. So what did the two podcast

going on. So what did the two podcast intersect in? Work becoming optional.

intersect in? Work becoming optional.

AI slashes the cost of living.

Productivity explodes. Income decouples

from jobs. Humans work by choice, not necessity. You don't have to believe it,

necessity. You don't have to believe it, but you do have to do it as an investor and think about it. And here's the chart of Tesla. Do you really want to be short

of Tesla. Do you really want to be short something that looks like it's about to explode? And again, another chart like

explode? And again, another chart like Corning, like Marll, like all of them.

basically from 2021 when PMI's peaked did nothing. You could overlay Marll's

did nothing. You could overlay Marll's chart on this. This thing looks like it's about to explode in a year where humanoids and robo taxis are going to become a major story. And by the way,

aside from Tesla, you've got SpaceX, which was just valued at $800 billion, largest private company. He built that company as well. And this was a really

good uh post that went out last night that I think again people should go because it gets back to my point about the government is supporting this.

There's no question SpaceX at 800 billion is the first open acknowledgement of reality that has been true for years in the United States is no longer the sole author of its technological destiny. It is now

technological destiny. It is now co-authoring it with a single private actor. The US can no longer project

actor. The US can no longer project force without SpaceX. Again, if you guys don't realize the importance of SpaceX

and satellites for military, for Starlink, which I use in Maine, this is critical. Everything related to this, he

critical. Everything related to this, he controls the most strategically important communications backbone on Earth. This is a huge advantage,

Earth. This is a huge advantage, including the fact that he has X AI. Joe

Rogan had a podcast this week with Jensen Yuang. again worth listening to.

Jensen Yuang. again worth listening to.

He talked about the kinetic AI as well.

Uh predicted a future where every country, every person will have excellent AI running directly on devices like smartphones without relying on the cloud. Again, that same visual I gave

cloud. Again, that same visual I gave you, not in the cloud, the LLMs. Go to the VA's, the rise of the VAS, the kinetic AI that controls physical robots and machines.

And again, here's the visual for that.

That's what's happening now.

My buddy Adam Parker put out a great report at Trivari Research. I heard him on a compound and friends podcast. I

asked him to send me over the the report. Uh highly highly highly

report. Uh highly highly highly recommend. He has this uh thing he calls

recommend. He has this uh thing he calls broken compounders. The reason I want to

broken compounders. The reason I want to bring this up is I want to bring this up for people focused on the growth side because I think this is critical. We are

leaving software. Forecasting three

years out in a world where AI is accelerating is impossible. What we've

seen with Salesforce.com and Adobe this year is not their earnings collapse.

This is not Amazon beating the retailers. This is a rerating. This is

retailers. This is a rerating. This is

taking companies built on code and saying, "You no longer have a monopoly.

You have competition. The competition is coming from Claude Code. It's coming

from Cursor. It's coming from individuals sitting in Africa and Germany and Vietnam and the Philippines and Brazil. They can build their own

and Brazil. They can build their own apps now. They can build their own CRM.

apps now. They can build their own CRM.

Small startup businesses will never use Salesforce.com. That wasn't the case

Salesforce.com. That wasn't the case before. If they got bigger, they would

before. If they got bigger, they would use it. You're going to have so many

use it. You're going to have so many small startup businesses using AI, they will build their own CRM system using AI. So, the growth where you used to get

AI. So, the growth where you used to get the growth has changed. He's talking

about broken compounders, stocks that fell 30% in one month and then basically had been up 100% over five years before.

This was a big year for that. He

highlights it in there. Most of them were midcap stocks. The majority of them were growth. You saw a lot of these. So

were growth. You saw a lot of these. So

these are names that collapsed 30% in a month, had compounded 5 years, 100% or more. So a lot of these they were

more. So a lot of these they were earnings related. They had high pees.

earnings related. They had high pees.

And again, you're going to see a lot of these going forward mainly in the technology sector. Most of them were in

technology sector. Most of them were in the most expensive side. This is the danger that comes in a in a in in a market with high multiple high-flying stocks. This identifies a symptom.

stocks. This identifies a symptom.

High-flying stocks are crashing and failing to bounce. The VA paper, my paper, bits to Adam's diagnosis the disease rerating due to capital rotation. The old thesis, the cognitive

rotation. The old thesis, the cognitive error. You are at risk if you are

error. You are at risk if you are heavily weighted to these companies, these software SAS companies. This is

not the end of SAS companies, but this is a point that if you're going to try to step in and buy these things, unless they're at the right multiple, these are becoming names that you're going to have to do based on growth rates, unless their multiples get into

the value category, which I don't see happening. So remember, the new

happening. So remember, the new compounders, the body of AI, the senses, the eyes, the nervous system, the muscles, arbitrage the gap. We're in a period

right now where you want to be wary of those software companies. You want to be wary of anything built on code. This

goes to Visa. This goes to Mastercard.

Remember, crypto is coming right behind it. I'm not doing a ton today on crypto,

it. I'm not doing a ton today on crypto, stable coins, AI agents. All of this compounding is going to disrupt established businesses which have had a monopoly. And if their business is built

monopoly. And if their business is built on code or they have high profit margins, they are at risk if their growth slows down. This is a very unique period in time. Here are the

semiconductors up 55%. Here's the

software companies up four. This is

since during the May the rally from May.

The S&P is up 22%.

I think you have to be wary of these.

And remember Adobe and Salesforce.com were not compounders in the fact that they didn't gap down 30% in one month. I

think the worst month for Salesforce was maybe 12%. This is death by a thousand

maybe 12%. This is death by a thousand paper cuts. Vibe coding is the chat GPT

paper cuts. Vibe coding is the chat GPT moment for code. That is what happened this year. here. And in my opinion, as

this year. here. And in my opinion, as someone who uses this stuff, vibe coding wasn't a word before this year. You all

know vibe coding or you've probably heard vibe coding. Vibe coding basically means you can sit down and create an app and you can sit there and just create ideas, stream through things, go throw

it, use cursor, use whatever, use claw code. The reality is this is the first

code. The reality is this is the first year of it. This is going to get worse going forward. The thesis is that AI is

going forward. The thesis is that AI is eating software.

high margin recurring revenue business models of traditional software companies are at risk of deflationary pressure from AI. Now what percent of those are

from AI. Now what percent of those are US companies?

85 to 90% of these codebased companies are in the United States. So codebased

companies which I break down in the MSCI because the MSCI is built on code. Now

approximately 15% of the MSCI world index is built on code. Now, what this does not include is a lot of the bigger companies. So,

Nvidia is obviously not included, but you're not seeing Google in here, uh, or you got Google here. You're not seeing Apple in here. You're not seeing the bigger companies, but you have a lot that have won that have been built on code. I believe there's competition for

code. I believe there's competition for these companies. And this is one of the

these companies. And this is one of the reasons why I like MSCI World X the US over MSCI US. Uh, I believe this reweing is not about the US exceptionalism. It's

about code now being completely commoditized and ubiquitous and the fact that code has dominated and led to this.

This is the other side of it as I've talked about. So you get a rerating on

talked about. So you get a rerating on MSCI world, you get a transfer from the application software and what you get is movements into the physical world. You

get movements back into the AI physical that is something more related to the rest of the world. commodity companies,

uh, sensor companies, optical companies, things that are made in various places.

Corning is just an example of a US company. There's German companies,

company. There's German companies, there's Swiss companies, there's French companies, all types of companies which don't make up a high percentage, but they will benefit significantly. Back to

uh, Aaron Ross, Andrew Ross Sorcin, who wrote the AI boom and the spectre of 1929. I bring him up because he did an

1929. I bring him up because he did an interview with Dario Modi which you can get on YouTube right here. Absolutely

worth listening to.

Here are the highlights. This is an honest thing from Dario Modi. And he's

asked about the bubble in case you didn't know it. Anthropics revenue

growth and this is the the need AI demand is compounding at a rate no one has ever seen. So one of the reasons that the buildout and this is him explaining why the buildout needs to happen the way that it is. So, if you're just saying it's an AI bubble, because

this has never happened before, you've also never seen a company that went from 0 to 100 million in one year, 100 million to a billion in two years, and then a billion to 10 billion in year

three. 10 times compounders, this has

three. 10 times compounders, this has never happened before. So, the 10 time compounding for three consecutive years hasn't happened. So, the reason they

hasn't happened. So, the reason they must spend tons of money going forward is because you have to build the compute years before the revenue arrives. You

don't have a choice. That's the way AI works. He admits there is a cone of

works. He admits there is a cone of uncertainty where it depends on getting the timing right. And it's how much money you're willing to do in capex versus verse how much revenue you're

going to do. If they buy too little, they can't serve the demand. They lose

to their competitors. If they buy too much, the business can die. This is the bet. And this is why everyone talking

bet. And this is why everyone talking about the AI bubble in effect is true.

there is there are going to be uh companies that either fail or just don't get the revenue they need. They'll be

bought out, their hardware will be bought out, whatever the case is. And so

maybe OpenAI isn't worth what they're worth. But the problem is they have to

worth. But the problem is they have to do this. So rather than say this isn't

do this. So rather than say this isn't going to work, I've gone through everything that's coming. You can't fade the whole thing. Certain companies,

especially the hyperscalers that are spending this, they will they are running a risky game. He covers all of this. It's worth a listen. also goes

this. It's worth a listen. also goes

through the vendor financing says why this is a necessity I'm not going to go through that he did say one company is yoloing it and I asked who was he referring to and

open AI why it's taking it the risk in one if revenue growth slows even modestly the companies that overbuilt compute could face existential financial stress this is the risk and the reason why if you

ask me the one most at risk is open AI I'll go through part of the reasons why he goes through the AI AGI race. He

basically says compute is the ultimate weapon. In the end, the winner is

weapon. In the end, the winner is whoever has the most compute, but if you get the timing wrong, you can bankrupt yourself. Some players are yoloing it or

yourself. Some players are yoloing it or yoloing that risk. This was an article that came out uh from Frontier Foundry, a software engineer there, uh Sultan

Mey, who runs it. I've known a long time. highly recommend reading uh this

time. highly recommend reading uh this basically going through the code red story in there and this is the reason I like this is because it's an honest assessment. Give me a four sentence

assessment. Give me a four sentence summary of the article to the LLM comparison. Jumal Google's Gemini 3 has

comparison. Jumal Google's Gemini 3 has overtaken GPD5 on key multimodal reasoning and benchmark tests triggering OpenAI's internal red code. I use Google Gemini now 70% of the time for all of my

research and the stuff that I do for this. I still use chat GBT mainly for

this. I still use chat GBT mainly for legacy purposes and conversation. I do

like the the interface more than Google Gemini, but the reality is I'm using it for the more important things. I'm also

using Anthropic far more at this point.

I use Perplexity for a lot of my uh earnings and stock related stuff.

Basically, Gemini, as far as I'm concerned, the analysis here that OpenAI is falling apart is true. Falling

behind, not apart. If you want to do it this way, I think this is a good way to do it. And I think there's an important

do it. And I think there's an important side here, which is OpenAI is losing in these categories at this point. Now,

could it jump back over? Yes, this has gone on before many, many times. Uh it

is possible that they could jump back this way and get in front of some of these with their next model since the most recent releases were these two uh guys and they did their uh Chat GBT5 during the summertime. So when Chad GBT6

comes out, maybe they leaprog back and uh ahead. Uh there's various ways it can

uh ahead. Uh there's various ways it can go, but the enterprise strength has clearly been done by anthropic. And if

you listen to Dario Modi, he goes through this. That's the bet they made.

through this. That's the bet they made.

Um I actually cancelled my pro chat. You

got Alon Mus. I didn't mention Grock. I

am also using Grock more these days. I

just want to bring this up because this guy is all over the place and I'm going to use him as an example. Basically,

he's fading Tesla. He's fading Nvidia, Palunteer.

He said Bitcoin is worthless. And he

talks about passive investing being the only reason these companies are higher.

The reason I bring this up, I can't stand even reading this stuff, uh, but he's picking on names that I think are going to be the best performers next year. So, uh, I think when he goes at

year. So, uh, I think when he goes at Tesla and he goes at Bitcoin, he's on the other side of the trade. So, what's

the difference? I'm going to get back into this concept of thinking in bets.

If you didn't buy the book for your kids for the holidays, you still have a little bit of time. I'm supporting Andy Duke on this, but I'm also supporting this because it's going to be part of my video in terms of the way I approach it.

You have to think in bets, especially when it comes to AI. The world has changed dramatically and you have to have a whole new mind. This book is another book you can buy. Uh this came

out I think in 2005. I read it uh just after the great financial crisis.

Phenomenal book. The reason it's phenomenal is Daniel Pink is making the argument that we are leaving the conceptual age or I'm sorry we're leaving the information age where there was left brain dominance. Now when you

say the information age the argument you're he's making is that you're moving away from math you're moving away from certain things uh the bankers and

lawyers all of these things that he mentions in there will be commoditized that's where we are with AI and what will be left are human- centered creative integrative thinking this really gets important because it's about

creativity empathy design I will argue that it's also about entrepreneurship it's about adaptability it's about the ability failing and getting back up and being an entrepreneur saying and

repeating the same mathematical view that you did with the big short which was not some big call but it was all mathematical. These are the reasons why

mathematical. These are the reasons why his linearbased approach to the market cannot work in an exponential AI world.

If you're falling into the trap of looking at the numbers this falls into the trap that people have made in trying to figure out why we haven't had a recession. This is still out there.

recession. This is still out there.

Every recession since 1960 has been preceded by a material downturn in this ratio of the LEI versus CEI. I've done

so many frigin things on this over the course of the year about how this is a useless statistic because of what it measures. It's not part of this world

measures. It's not part of this world anymore. Manufacturing hours work,

anymore. Manufacturing hours work, credit conditions, orders, surveys. It's

a different world. We have left the world of manufacturing and the industrial economy which was was built for which GDP was built for. We've

entered a world of productivity, efficiency, and exponential innovation.

You cannot use these old tools. So, you

have to think differently. Here's the

old system. This is where Michael Bur and the LEI are trapped. It worked

phenomenally well for this side. It

doesn't work for times where companies can go from zero to 10 billion in revenue. 0 to 10 billion in revenue in

revenue. 0 to 10 billion in revenue in two and change years. You can't have that going on over here because you needed physical structures. This all

changed in 2007 when the smartphone came out. Every dollar that's been made

out. Every dollar that's been made investing has been made on the Mag 7, on Amazon, on all of these things because they were part of this. This is why again I went to Singularity University.

It's still time. If you want to get your brain to change and you want to learn something, go to the executive program at Singularity University. Peter Deandis

set it up. I references moonshots. I

went there in 2013. Here is a just a preview of the website. Um, for people who don't get the research that I do for 22V, which will have a whole bunch of

different reports and things in it.

Also, I'm going to build the video series which will come out which will help you to do things like this in terms of drilling down. How do you find Corning starting from a podcast or from

a phrase that you hear? How do you get it down to an investable name? Now what

will go on in the master class in terms of the videos and again this was done you can see this really cool design this was done in nano banana again it's going to be focused on thinking in bets and

make your bet I'm going to try to do this with two powerful philosophies constantly thinking in probabilities using AI not to find answers but to continually increase the probabilities

of success using technical analysis using things on the investment side but to get there it's all about being an entrepor reneur and it's about starting small at the very beginning. It's

literally a lifestyle change. So, think

of this as a work style change. Finish

up Bitcoin building a bottom here. Maybe we come back down 80,000 again. I don't think so, but you're starting to get in there.

Uh Robin Hood, this was this should have been before the last slide. Um this is basically what I'm I'm I'm using. This

is from a Robin Hood recent uh presentation for kids sitting at home who want to learn the ability to invest, the ability to use AI. Robin Hood has

been dominating the financial markets.

It has been dominating everything. We

are moving to tokenization next year, which means the speed of things is going to increase. If you're not using AI to

to increase. If you're not using AI to find ideas, to deep do deep research on ideas, you're playing from a place of weakness. It will be connected to crypto

weakness. It will be connected to crypto next year. That's why I brought it in

next year. That's why I brought it in this. So, just pay attention to it. For

this. So, just pay attention to it. For

tokenization, you're going to need more Ethereum. Here's the breaking line. I do

Ethereum. Here's the breaking line. I do

think when we trade above here for 2 days, that will be the beginning. If you

remember what happened here, we had a similar breakout there. I think this trend line is going to be the key to all crypto. I think it's coming soon. I'd be

crypto. I think it's coming soon. I'd be

paying attention on it whether it happens next week or the week after.

Final few slides. Tokenization uh report from Black Rockck in the Economist.

Again, tokenization is a major theme for next year. So add it to the humanoids

next year. So add it to the humanoids and the robo taxis. The upgrade cycle PMI is going higher. Goldman acquires a Bitcoin structured fund in its latest

deal. And finally, I will be at this

deal. And finally, I will be at this event in Miami like I was last year. Uh

Pomp's going to be there as well.

Hopefully, we're going to be able to do our our weekly video from there. Uh it's

been a great year, guys, in terms of going through it. I will be with you, I'm sure, the next few weeks. Uh

hopefully this helps. Go to the 22V website constantly see when everything is done. Reach out to the salesforce

is done. Reach out to the salesforce there if you need anything in terms of what I went through. Uh, subscribe and thanks again. And see you next week.

thanks again. And see you next week.