AI叙事下的市场波动：甲骨文与OpenAI新闻如何冲击股市

All right. Hello everybody. Welcome back to another episode of the market close.

It is Friday, December 12th, and we are ending out the uh second week of December.

Ugly day. Ugly, ugly day.

S&P 500 here right here down about 1.12%. I think the worst of it was 1.15.

And you've got the slightest of slightest bounces, but nonetheless, still a pretty nasty day.

Nvidia right there, 175, that's down 3% on the day.

Palenteer is down 2% on the day. Hoodie is down 3.2% on the day.

All these are up a little bit from their absolute lows, but nonetheless, they are still down.

Broadcom absolutely disgusting. It got to 355, it's at 357.

We'll talk more about Broadcom, but that's down 12% on the day.

was the first domino to fall, which ended up cascading into a variety of other AI headlines that ended up it's not being pretty for the markets.

Um, but Broadcom got the brunt of it today.

Netflix holding on to its green, Bitmine's down 8% because Bitcoin and Ethereum are down, Bloom Energy is down, Iran is down, Amazon is down, all of these names, anything that is in the risk on part of the market.

Bitcoin at least holding above 90, it did get to 89.5 is down.

So, what happened today? What happened today?

Well, I think at around 11:00 a.m. Eastern this morning, there was a headline that came out that uh OpenAI would be delaying a data center buildout with Oracle.

And well, that made the market a bit upset because the idea is that OpenAI can't pay for any of these data centers.

A lot of this money is fake.

There's $300 billion of remaining performance obligations from Oracle never really existed in the context of OpenAI's potential deal with them.

And so the market's upset about this idea that we're never going to be able to get uh this type of money and so we shouldn't be giving Oracle this type of a premium that bleeds into all the other AI names.

Oracle comes out an hour later and says three things.

Number one, site selection and delivery timelines for OpenAI data centers were set jointly with OpenAI.

There's no delays to any sites and all milestones are still on track.

Now this was somewhat appeasing to the market.

The market started to recover a little bit when this news came out.

problem is it doesn't necessarily make the argument that uh we're going to immediately begin the performance obligations from this openi contract in 2026 and I think that's the issue right here.

So you notice how they word site selection and delivery timelines for openi data centers were set jointly with open AI.

Basically, they're saying, "Hey, if you guys think there's a delay, then yeah, maybe you can interpret it as a delay, but Open AI and us always had, for example, 2027 in mind, not 2026."

So, they're kind of saying like, "Yeah, it is a delay in the perception of what you guys thought it was going to be, but we never said it was going to be 2026.

At least I don't think Oracle ever said that.

And so, we always thought it was going to be 2027, which means now that the news is coming out that it's not going to be 2026, it's going to be 2027.

You guys want to freak out. That's not our problem. That's your problem.

I think that's essentially how Oracle framed this argument.

And when they say there were no delays in place, they're basically saying, "We never thought it was going to start that early."

And if you guys thought it was going to start that early, again, that's not something that we can be concerned about.

Oracle right here, their credit default swaps got to a fourmonth high today.

They're up 200% in the past four months.

That stock's still down 4.72% on the day.

Same thing with Broadcom.

So, you had Broadcom that kicked it off down 12%.

The reason Broadcom was down, unfortunately, they also said that their exclusive 10 gawatt ASIC data center agreement with OpenAI is not going to start in 2026.

Market wasn't upset about that or market was pretty upset about that last night, ended up being more upset about that this morning.

You're one-two shot that'll give you a red day and that's obviously affecting the rest of the stock market.

Nvidia being a big name that's going to get hit because it's center in the AI trade as that's down 3% on the day.

S&P 500 down 1.13%.

The crazy thing about this Tesla actually holding green.

So Tesla was at 463 came all the way down to 445 now it's back to 457.

So uh Tesla which tends to be the most high beta MAG 7 actually holding on to this momentum um in the midst of some of this AI collapse.

But I the ironic thing about this is that we were at alltime highs yesterday after hours.

And so I think the whipssaw of emotions and price trajectory and uh essentially price action that we've been used to over the past couple weeks that is continuing.

We had a little bit of reprieve over the past week and a half as we got back from Thanksgiving where it wouldn't be that big of a deal or we thought it might not be that big of a deal and it came right back.

BMR down 9% on the days.

I think the takeaway here and you know we'll talk more about some of this stuff as we get through the show but uh the takeaway is the market is fragile and the market fundamentally gets scared every time there is a hiccup on the AI narrative and today we had two hiccups there's there's no if ends or buts about it we had two hiccups with this Oracle headline whether the headline is reasonable or not if it changes the entire thesis or not which I don't think it does but there were some hiccups and that's why the market got upset and uh I think that's why we're seeing some of this pain.

Okay, Trump is live right now as well.

We're getting some quotes from him.

Uh, let's pull up some of these quotes right here. Trump says, "I'm leaning towards Worsher Hasset to lead the Fed."

Trump says next Fed chair should consult with him on interest rates.

Trump says interest rates should be 1% or lower a year from now.

Um, I don't know if we're going to get 1% rates uh next year, but there you go.

So, Trumpet is Trump is uh uh narrowing the the search down to Hasset or Worsh, which is kind of what we all thought, but Hasset or Worish.

I still think it's going to be Hassid might like Worsh, but uh Hassid might end up being the one who wins.

1% rates, that's what Trump is saying right here.

And yes, Trump asked where he wants interest rates to be a year from now.

Trump said 1% maybe lower than that.

I mean, obviously, Trump is going to say whatever he wants to say, but I don't know if we're actually going to get 1% rates or lower.

we are I mean maybe this market is going a lot higher and if that results in a crazy bubble it is what it is.

but if rates are going lower you might have some money continue to be put into the riskon area of the market.

by the way speaking of money that needs to be allocated money market funds we didn't get to check it yesterday because we had broadcom earnings kind of a nothing burger 1.19 billion 1.19 billion

I'm buying so many houses if I have 1% rates yeah hopefully that gets mortgage rates down to at least 3 or 4%.

If not lower.

Uh but then again, you know, it's going to be hard to see if that actually plays out.

1.19 billion into money market funds that came in over the past week.

Last week it was 85 billion.

So again, nothing too crazy, but that is what we got.

All right, let's hear from uh Oswth and see his perspective. He's the dean of valuations at NYU.

This is his current thoughts on the state of the AI trade and where the market is going.

>> Awesome.

And I think that's at the basis for why there's so much disagreement.

It depends on how plausible you see that pathway is.

Give you an example.

Take Nvidia at a $5 trillion market cap.

If they can maintain their existing nosebleleed margins, 53% net margins, return on equity in the triple digits, they can get there with 600 billion in revenue, which is not a big deal.

I mean, they're at 170 billion already.

With 20% growth, you can get there.

So the people who are upbeat about Nvidia are seeing a plausible pathway there if they can if they can do that.

The question is is that probable and that's where the disagreement comes in.

And I think that's why it's got to be personal.

It's not something that you can hear from other people.

You got to look at the numbers and say I can live with this.

I can make this assumption and be okay with it.

And every one of these companies I think has a plausible pathway to get to the market gap they're at.

And you know, as long as that's there, you're going to get disagreement about these stocks.

>> I think what's really provocative from you today in terms of your notes to our producers is that you put forth that this is not the bifurcated market that some would paint it to be.

That the remaining 493 are also trading in your mind hefty multiples of earnings.

Explain.

Mhm. I mean, you take the seven out of the S&P 500, look at the rest of the market.

It's trading at 22 to 23 times earnings.

And you could argue that the remaining 493 companies have much less rosy operating metrics than the Mag 7.

They don't have the kinds of margins and the growth that the MAG 7 have.

So, when people talk about this being a bifurcated market, they act like the Mag 7 is essentially the only they're the only ones that are overvalued.

The rest of the market somehow is fairly valued.

I don't think that's true.

This is a richly priced market across the board.

If you fact the differences in margins and growth, I'm not sure the remaining 493 are bargains relative to the Mag 7.

They look pretty richly priced to me as well.

>> What looks undervalued to you? If anything, I presumably you've thought about that.

If anything, I presumably you've thought about that.

>> It's I mean that's why I said coming to a conclusion that the Mag 7 are overvalued, the rest are undervalued is going to be tough to do because everything's been pushed up.

It's like a rising tide.

And with the Mag 7, you have this added oomph of a high growth plus high margins at scale.

I mean, that's the amazing part is they're growing when there are 150 billion, 200 billion, $250 billion revenue companies.

That's I think you know almost unprecedented and the market is factoring that in rewarding these companies.

Whether they can keep doing that becomes the area of disagreement because I think that things only get rockier and more difficult from this point on.

You saw that when Google announced, you know, announced that these new chips that compete with Nvidia chips. You saw the sharp sell off in Nvidia.

I think you're going to get more shocks like that to the system and the question whether the story can hold up.

>> Huh. What does your own portfolio look like these days? You you've never really been shy in in sharing the moves that you make. Could you could you do it again?

You you've never really been shy in in sharing the moves that you make. Could you could you do it again?

>> I'm down to five of the max seven.

They're there because they were all they've been there and they're not overvalued enough that I want to dump them.

I did you know sell off the rest of my Nvidia shares just recently and I think the la that was the last segment of Nvidia.

I've you know I and if it continues to go up all the more power to people who continue to hold in video Nvidia but for me I think I've gained enough from its from its rise that be greedy for me to hang in there say give me more

>> but do do you agree with those who say that 2026 is going to continue to see more dispersion among the mega cap stocks among the mag sevens that they're just not all going to trade like the monolith that they once did

>> and they have for about a year right even If you look at this year, they've been the max seven have moved in different directions.

This is not 2022 or 2023.

But if you could tell me what one of the max seven stocks did during a week, I could tell you what the other six did.

Now, if you tell me what Bitcoin did today, I can tell you what Nvidia did did.

I think you have a better shot of predicting how these stocks will do by looking outside of the Mac 7.

So, uh, the perspective continues to be by Oswth that, uh, everything is overvalued and, you know, he's been holding some of these maximum names for a long time, which is why I guess he's not selling some of the other ones.

Uh, he is out of Nvidia.

He does not believe this one is worth holding on to.

And the S&P 4, wow, Oaklo down 14%.

I didn't even see that.

And down two, Roblox down six.

Everything is really hard to find some safety.

I mean, maybe Netflix was safe.

uh Verizon, Tesla ended up being safe today.

Lulu ended up being safe, but everything else getting hit.

His argument is the S&P 493 are also overvalued and there's not a decent level of excitement there.

Now, uh we're going to listen to some other clips throughout the rest of the show, but uh I don't know if you guys have had a chance to listen to the new Tom Lee podcast on Prof on on Scott Galloway's thing.

I I think I played like a one minute clip of it in the morning.

I listened to the whole thing throughout like since I left the market open, guys.

He's not the most bullish and we'll we'll analyze it.

But if you listen to Tom and you really listen to what he's saying, he's not he's not bullish.

like he's he has this $7,700 target, but even the way he was answering questions about bubbles, concerns around AI, the way he's talking about the labor market, I mean, you guys can judge for yourself when you listen to it, but it's not the same bullish Tom Lee from a month ago.

I mean, fundamentally, I felt there was something different and in in many respects, he was agreeing with a lot of the arguments um that the podcasters were asking asking him about and that came from a much more bearish perspective.

So, I mean, he has a target, but he's calling for a 20% draw down on the S&P before we get anywhere near 7,700.

And so, I I I thought I just thought it was interesting to listen to how he's kind of framing it.

And we'll kind of go through it together.

But, nonetheless, you are seeing some more people start to ask deeper questions.

And unfortunately, on a day where Oracle ends up giving you this OpenAI news, it compounds and, you know, results in a pretty negative day.

And unfortunately, that's what we're kind of seeing on today.

So, we got the close in about 10 seconds.

We'll talk more about it.

We'll talk about some of this Japan stuff coming up next week as well.

We have a couple other macro headlines.

We got we're getting these Wall Street Journal quotes from Trump now about the Fed.

And so, we'll go deeper into that and analyze what that looks like as we get into the close.

All right, Friday, December 12th, 4 p.m.

Thank you everybody for being here.

The stock market is now closed and we have ended out the second week of December.

All right, the bleeding stopped.

Yes, hopefully after hours gets the bleeding to take a little bit of a of a pause.

But nonetheless, ugly day.

No way else to kind of frame it other than it was ugly.

Nvidia right there 175 down about 3.2%.

Nvidia now $4.25 trillion market cap down from about 4.7 earlier in the week.

S&P 500 right there 681 uh that was down about 1% on the day.

EOSE that was down 9% Oaklo down 15%.

Every energy name got hit based on the idea that Oracle and OpenAI were delaying these data centers and as a result of delaying these data centers uh the energy names are feeling some of the pain because the energy names were going to catch a bid because of the data centers.

I I mean I think that's the reason.

Maybe there was something else.

Obviously, a lot of these names are high beta, but that's why I think you're seeing a lot.

I mean, even ens we we talked about ENS on equity edge um like three weeks ago.

The stock does not have a massive multiple.

Even this one was down on the day $5 billion market cap given if it given it's in the energy ecosystem.

Robin Hood down 3% on the day.

Palenteer down 2.1%.

Paler still holding 180.

It was down to 179 at one point.

So, you did get a little bit of a rebound, but that is down from about 187 early in the morning.

Apple little bit of safeties today.

278. That is green.

Oscar was up about 5 percent.

There is now a 30% chance on prediction markets that we're going to get a shutdown.

It was like 10% last week.

So something happened throughout the day to get the prediction markets to think that we will get a much higher probability of a shutdown, which again is not the best news going into January.

So we'll see any implications from that.

UNH, that's up about 1.4%.

Grab did hold around 520 throughout the day.

Ended up falling 3 cents as we got to the end of the day, 515.

So that's down about half a percent, but it was holding 520 for the majority of the day.

The weed stocks uh still ended up doing well.

Aurora Cannabis up 18% on the day.

We did not get an executive order with the scheduling of marijuana as a schedule 3 drug.

Uh so that hasn't happened yet, but that still is getting the market to be excited.

Planet Labs right there.

That's up 4.38%.

And then where is Rocket Lab?

Rocket Lab is holding 61.

Up a good percentage for the week.

Can't be too upset about it.

Down from 65.

So down 3% on the day, but still holding 60, which is not too bad.

Disney, we had that big announcement with Open AI.

That one's basically flat on the day.

Verizon was flat on the day.

Uh, and then Tesla, which I still don't know why Tesla pumped.

I think the executive order on the federal mandate for AI standards probably is going to make **FSD** (Full Self-Driving: 特斯拉的全自动驾驶系统) more regulated and not have to go state-tostate to be able to get through unsupervised FSD.

And maybe that's the reason for the pump.

But outside of that, not sure why Tesla got a bid, but it was one of the only Max 7 today that got a bid.

Amazon as well ended down on the day 2.3% 226 on Amazon as we are ending out this market for the close.

Amcor was down as well.

I mean Amcor is a big supplier with uh Amcore down 6% with Broadcom.

So unfortunately Broadcom 11.43% Broadcom lost about 40 billion in market cap.

A lot of derivative plays.

Where did Google end?

Google 309.

Okay.

It was 308 in the morning.

305 at the bottom.

309 to jump to that one.

Still down 1% as well.

That's what we got.

That's what we got on the day.

Let me know if you bought the dip.

Uh if you took advantage of any of these stocks.

I bought a couple of these names in the morning.

Uh not massive massive amounts, but just a nibble on some of my kind of core positions that I care about.

Outside of that, and again, we'll we'll we'll go deeper and talk into it, but I don't know.

Today's price action was weird.

You had a lot of different perspectives come through and some of those perspectives argue that this Oracle stuff is stupid.

Other perspectives came through and said, well, this is going to be the beginning of something bigger in regards to the delay of data center construction, which theoretically could hurt a lot of these other names.

And you know, we we know that there's a bottleneck when it comes to power for data centers.

So, I don't think that's surprising.

But the deeper concern is like the way the market priced a lot of AI stocks today was that there's no demand for data centers.

It wasn't even like oh there's a logistical concern about how to set up the data center.

It's like oh no there's no demand which is not the case or at least that's not the evidence that we were presented today.

But yeah, as Happy Mey says right there, the market is very conflicted right now.

Incredibly conflicted.

And that's why I think we're seeing some of the uh some of the pain.

Uh, I see some Nvidia.

I see VG.

I see Next.

I see a little bit of Netflix.

Uh, I see some Hood.

I see SoFi.

VRT.

I see Broadcom.

A lot of people are waiting for 360, 355.

I mean, you got close to 355 this morning.

VRT.

Look at that.

Down 9% of the day.

Another data center play in um in liquid cooling.

I see a bunch of Nebuses.

I ran.

Did that close below 40?

That's Yeah.

Well, no, not below 40, but just above it.

Still down 8% on the day.

Rubric M.

Like, why was Rubric down?

Like, did the data center think have anything to do with Rubric?

They had a great earnings.

So, I do think a lot of these names unnecessarily took a hit, but they did indeed take a hit.

Bitcoin, right, they're 90,000.3 as that was down about 2% on the day.

Amit, do you think Tesla's good for covered calls here?

To be honest, I I don't know why Tesla moved today.

If anyone has a reason, let me know in the chat.

So, yeah, I would still be conservative, but Tesla breaking out to 500 by end of year or even that 480, 490 might be a little tough if the S&P is not helping it get there.

Going against the S&P today was nice to see, but then again, Tesla does this all the time and then it tanks the next week.

Like, it's just one of those names that has a lot of momentum up or down.

That's what makes the premiums amazing.

So yeah, I wouldn't get super aggressive and do like a 460 call, but like 475, I don't think it's too crazy going through by the end of the month.

**Rivian** (Rivian: 一家美国电动汽车制造商) was up 20% on the day.

Rivian ended up only down 12 up 12%.

Still a great day for Rivian.

Let's look at the heat map today.

Rivian, Lulu, couple of these defensives ended up holding on throughout the day.

This was what tech looked like.

It was red.

Microsoft, Oracle, Palanteer, Broadcom, Nvidia, AMD, Micron both were down five% on the day.

Financials held strong.

Healthcare held strong.

Google and Meta got hit.

Netflix held strong.

Amazon got hit.

Again, Tesla was one of the only tech stocks that was actually green along with Adobe.

Adobe got some love today.

That was up 2%.

And then Walmart, Costco, Consumer Defensives ended up holding a little bit of green.

Did Tesla move on the Rivian news?

I don't think so.

I think maybe a little weird if that would be the case.

Yeah, the Marco's like, "OMG, Rivian chip's going to replace Nvidia and data centers, right?"

Like that's how it almost felt Rivian moved today based on the the chip announcement.

But um yeah, I don't think Tesla necessarily followed that.

Okay, let's uh let let's let's let's pull up this Tom Lee argument uh on the podcast today.

I mean, I I do think there maybe it's just me, but and he still is bullish, but I do think there is a different context of how he presented his arguments.

And I'll let you guys be the the the uh the jury of this.

So, let's play this for a couple minutes and we'll get some perspectives.

But there is a the the point of what he's saying in this podcast, at least that's what I got from it is expect a downturn of 20% next year.

Like be ready for that.

That will happen.

It's almost like he's speaking that it's inevitable because we're getting frothy essentially is what I got from his argument.

And then you'll get a recovery to 7,700 which means we would have to recover, you know, much more than 20%.

But that's kind of the broader argument that I got from this podcast.

And do you think those would be the trigger of such a correction?

>> Let's start with the one that you just mentioned.

The stock market, you know, it we're up 16%.

So I think if we rally three percentage points, we'll be three years of 20% gains back to back.

Um, and it's actually a couple of you guys in the comments saying, "I picked up on how you're feeling."

Okay, so I'm happy the rest of you guys will hear right now, but I'm happy it's not just me.

There's a couple of you that listened to the podcast already and you're saying you picked up on that same feeling.

I think it's kind of evident the way that he's describing kind of what he thinks could play out for 2026.

More common than we realize.

Um, in fact, when we look at uh the last 65 years, you know, it's happened in 20 different I think it's happened 20 times in different countries um and multiple times in the US.

I'm sorry, 12 times.

It means a lot of good news is priced in.

I mean, of course, you know, stocks being up 20% a year, three years in a row, it's definitely pricing in a lot of good news.

So, to me, I do think that we have to consolidate those gains and and that's why I I think a draw down next year makes perfect sense to me.

But, uh because there isn't a lot of leverage in the economy, you know, household sector has not really borrowed money.

It's been expensive to borrow money and even margin debt, it's risen, but it hasn't risen parabolically.

So, it's actually essentially tracked S&P gains.

Um, so that's it's not like people are borrowing faster than the market's been going up, especially if you look at a 5-year um kagger.

So I I I would be in the camp that as long as the economy's holding up that draw down is going to be viewed as as a buying opportunity.

Um now on AI valuations it makes perfect sense for someone to say a lot of the valuations for AI are probably absurd because this is the nature of like of a of a exponential sector, right?

If we look at a industry that could grow parabolically for 10 years, all of the future value is in the latter half of those years, right?

So it's and then we're trying to discount that back to today.

And so stocks are going to look absurdly expensive.

And more importantly, investors make a common mistake, which is that they assume that the existing universe of companies are going to be the central cast characters over the next 10 years, which is not the case.

So, the reason valuations don't make sense today is that one of all the AI stocks, I'd say it's probably safe to say only 10% are going to be good investments.

Maybe it's even generous, maybe 5%.

I mean, when's the last time you heard that from Tom Lee?

By the way, Derek, thank you for being here.

Thank you for the gifted gifted 50.

50 of you guys just got gifted memberships courtesy of Mr. Derek Wall.

So, thank you, Mr. Wall, for those gifted 50.

But, I mean, again, when's the last time you heard Tom Lee say that only 10% of the AI companies today are going to be good investments?

That's being generous.

Maybe 5%.

And again, that's a totally reasonable opinion.

And I think that's actually, one could argue, a logical opinion.

There's a lot of these AI stocks that are just like stupid.

but you never heard it from Tom in that much of an explicit way.

And and and as you go through the podcast, you start to recognize a little bit more cause for concern that he's putting out there, but he's also kind of saving himself, right?

Because he's like, "Oh, 20% draw down, but then we recovered to 7,700."

It's like, I love you, Tom, but you're playing both sides here, right?

I mean, we have a last time we had a 30% draw down.

The only reason we recovered in five months is because Trump immediately pivoted the whole tariff stuff.

I mean, if if it's not like a light switch type of buy the dip opportunity like it was in April, which was, you know, that doesn't happen that often to go down 20% and then recover 40%.

I I I mean, it begs the question like how likely of a scenario do you think that is?

I I'd rather believe in a situation where we don't have a massive draw down, but we also don't have a massive upside case.

So maybe the S&P is 5% next year on on on on that perspective, but to go down 20 then go up 40, I don't know how plausible that is to to believe as well.

>> And of course, um there's going to be a new emergence set of new players.

And in fact, the economic model might change, but it doesn't mean it's a bad investment.

And and we we've highlighted this as generational trades in past reports.

For instance, like if you look at the internet um if you bought the internet basket in 99, okay, so you bought it near the peak and you held it to today, you actually still outperform the S&P 500 even though 99% of the stocks went to zero.

So it wasn't it was a bad investment if you tried to pick a winner, but it wasn't so bad if you held it as a basket.

So, I think AI, it's probably going to be fair to say 90% of the stocks are going to be do way worse than people expected.

They were too optimistic.

But I think as a basket, it's probably going to outperform.

That all makes sense to me.

I'm with you.

But it seems to be a little bit more nerve-wracking when we realize that a lot of the AI companies are the largest companies in the world.

It's the big tech companies.

I mean, I think Google is an AI company at this point or an AI stock.

Meta, Nvidia, I mean these are the largest most valuable companies in the world.

Um, and the market really depends on their their performance.

So when I think about the idea that you know many of these companies and the and the expectations that have been pinned to the AI cycle, the fact that that could affect some of the largest most valuable companies in the world where we're seeing the highest concentrations in those small companies that higher concentration than we've ever seen uh in history to me that makes it scarier what you just said.

So, I guess my question is, do those companies, do the big tech companies, the Magnificent 7, do they count in your analysis of AI valuations being too high and the possibility that perhaps we might lose out or that that the value won't actually show up for many of these companies?

I might even just add to your concern because there's a again I might even just add to your concern like I'm I'm just I'm just presenting what I heard an hour ago.

I mean, you got Tom Lee saying, "Yeah, you're right.

And here's another reason to be concerned."

I I'm just saying it's obviously we're trying to crowdsource perspectives and understand this stuff, but this was not the type of bullish sentiment you've been hearing from him throughout the entire year.

because there's a lot of capex here too.

So that these are, you know, a lot of the mag seven used to be asset light businesses.

you know they the remarkable uh equity you sort of rent seeeking model of them was that they could create growth with very little spending.

I mean R&D spending was there but really was not there but today as you know uh AI is extremely capital intensive and it's energy intensive and it's only justifiable if it's replacing real work somewhere else then you can justify because now it's creating assets to replace future opex you know I'm going to give you a spin about what's happening that is not disagreeing with what you're saying but it's probably observing a change in the reality.

Okay, which is tech companies are becoming a bigger part of our life.

Um so naturally they're going to have a larger share of spending.

Uh, by the way, we wrote about that um in 2018 that in if you go back to 1930 and you just use simple demography, okay, population tables, whenever the population growth rate grows faster than the prime age workforce, which means you have compounded labor deficit, you've always had a technology cycle.

That was 1948-67 and 1991 to99.

In both of those periods, the population growth rate was growing fast, which is demand faster than worker supply.

And we entered the third epic or era of labor shortage, which started in 2018 and it's going to last to 2035.

So then technology spend is necessity because you don't have as much labor available.

So there's going to be less wage spend.

Yeah.

Now if I substituted the word and called this um instead of the word banks uh tech companies I called them financial institutions.

We would not be saying there's a financial institution bubble because for every level for every unit of GDP growth, there's a unit of financial spend.

I mean, it's literally the other part of the ledger.

And in fact, the financial industry has all circular spending.

I mean, think about this.

Real estate is valued as a separate asset, but every company needs real estate just to run a business.

So, why are we valuing real estate like in a GDB sense?

Real estate should be an interim product, not a final product.

Um, so I think tech is becoming so central to the economy, especially because of labor shortage, that we're when we see tech intensity growing, people are flagging that as a bubble, whereas I'm actually just pointing out is it's it's actually out of economic necessity.

But it becomes a bubble if the multiple we're applying to the tech streams don't justify higher valuation.

I think tech earnings are probably more valuable than Costco for instance.

Yeah, I agreed.

Or Walmart, right?

But do you know Walmart trades at 37 times board earnings and Costco trades at 50 times?

So Nvidia trading at 27.

I mean is that is there a bubble in Costco and Walmart?

Because Nvidia is at 27 times earnings.

>> 100%.

And we've looked at that.

So uh you can listen to the podcast yourself and as you can see right there he's still making the argument right there.

I'll put the link to the podcast in the chat that obviously we are not necessarily seeing the the bubishious behavior that the street is expecting or or even perceiving because of for example Costco and Walmart not being the most important financial institutions or companies in the country very important countries but one could argue that the tech companies are probably more important but a company like Nvidia which is at the center of the AI trade is not trading at a euphoric multiple given it's trading at uh 25 times next year's earnings so I mean look there is an It it's kind of the takeaway I got from that is a hedge to both sides.

There is an analysis on why for the first time ever I've heard Tom say 90% of the AI stocks are going to zero which look there's a lot of stupid crappy dumb names that just put AI in the ticker and they say like they're an AI company.

They're not even growing.

Like yeah those are probably not going to be some of the best companies in the world.

But it's the first time you heard Tom say 90% of the AI cap bucket of stocks.

I mean, he's essentially comparing it to the internet stocks, which 90% of those went to zero.

And we got Amazon and Google from it, which why it ended up outperforming the S&P because those companies were just so transformational.

And I'm assuming the most important transformational AI companies will probably outperform the S&P over the next 10 years.

But you are seeing the first resemblance of him making that claim that like a lot of these companies probably are invaluable.

And then sort of the second argument he's making right here is that um this idea of consolidating the gains after a three-year bull market which is entirely reasonable and I don't even think there's like a problem in in in this argument this in this chain of thinking from a like logic perspective that if you have three years back to back of 20% let's say we end this year at 15 16% whatever it is um those are a massive three years we have to consolidate the problem is his argument and this is the part where he's losing me here is that let's just Say S&P ends the year at 690.

Okay, 690 losing 20% would put us back at um 552.

Okay.

552 on the S&P.

So then to get to 770, we would have to go up Yeah.

39% 39.4% 40% to get to his level of 770 7700.

So his outlook and again we're going to get a lot more outlooks from a lot of these other banks.

Not all of them agree with him, but Tom Lee is saying we're going to go down from 690 essentially to 550 and then somehow someway we're going to rocket ship back up to 77,700.

Yeah, it it would be a repeat of 2025.

The difference is, I mean, look, I I don't know how I'm probably missing some some some economic history here, but how many times have we gone 30% and recovered within two months?

And the reason we went down 30% was because of Liberation Day.

The reason we recovered is because Trump said, "Buy the dip."

Like, there was an obvious catalyst to the downside and then that catalyst immediately reversed to the upside.

It it was an anomaly.

It created one of the best buying opportunities in history.

But like to expect that to happen twice, I don't know.

And I don't know why that would even have a chance of happening, right?

Like no one expected the liberation day thing to be as insane as it was.

No one expected the recovery to be as insane as it was.

Do do we get something like that happening again?

But I don't know.

I I I think it sounds kind of off that we're going to go down 20 and then go up 40.

And I think he's trying.

And by the way, if we go down 20, I mean, where where the hell is Ethereum go?

If the S&P is down 20%.

What?

Ethereum's down 50,500.

If that Bitcoin obviously gets destroyed in that scenario, and if that does happen, obviously that's the time to buy.

Like if we have a 20% drought on ever on the S&P, you you know that forget extreme fear.

You know that that is the moment you buy.

The problem is to try to predict that then we'll go up 40% and recover all those gains and go higher, especially if you have more AI bubble fears that can compound on each other.

I just don't know if that narrative makes sense for next year.

I think there's other ways you can make a narrative for what happens in 2026, but I'm not going crazy, right?

Like that just sounds a little not believable.

And it's the first time and it's ironic that this podcast dropped on a day where this whole Oracle OpenAI stuff came through.

Um but but it is a bit ironic that this is kind of his frame of analysis right here and then you know you're getting these concerns about data science.

So look I think we have to obviously do a lot more work and in terms of analyze what's going on.

We're gonna this Tom Lee is just one person.

I'm gonna try to get a lot more opinions on seeing some of what these other companies are saying.

I want to know what the consensus S&P target is for next year because I think the highest I've seen is 8,100.

I think Everor at 9,000 in like their Super Bowl bull case and then you know the the the the I guess the lowest I've seen so far is like 5700 6,000 but we haven't had a consensus view.

The consensus going into 2026, by the way, or 2025 was 6750, which we're probably going to close above if we, you know, hold these levels and don't fall another two or 3%.

Because we're at 681, but the analysts kind of got it right going into 2025 because I mean, we're relatively at that same level unless we get back up to to 7,000.

So yeah, I think these are two things that we definitely need to think about.

Would the Supreme Court tariff ruling be the reason for the draw down?

Um I I actually think that would be the reason for the uh upswing because the market doesn't like tariffs.

The Fed definitely doesn't like tariffs.

Inflation potentially might not be that big of an issue.

Although there's two push backs to that.

Number one, the administration came out yesterday.

I think uh Bent said that they've got or it was either Bent or Hassid said they have plans to deal with the Supreme Court if if they do overturn overturn the tariff stuff.

So they're going to try to institute the tariffs in a different way.

And then the other argument here is that I mean to the draw down point it's like if the market has uncertainty around tariffs that's not a good thing.

Even if Trump is able to finagle his way into still implementing the tariffs which would also create more uncertainty because then it would probably be questioned from a legal perspective.

So yeah, could it result in a draw down?

100%.

Could it also result in an upswing and then maybe later the market because they're just so scared about what's exactly going to happen that results in a bit of a draw down?

Yes.

But again, no one knows.

And I think that's why it creates a lot more uncertainty.

Uh let's listen to his perspective on the labor market because that is also a pretty important issue right here.

And this is what he had to say about where labor will go.

One of the other bare cases for most people is that if we have this AI boom, but people lose their jobs, that's not necessarily going to be the healthiest for the economy.

We're in this labor shortage cycle from 2018 that will last through 2035.

And I I just you can't avoid the catastrophizing around the destruction of the labor force from AI.

Do you still believe we're going to be in a cycle of labor shortage even with AI?

I vacasillate because there's times where I'm I like when I read a book like The Coming Wave, I panic and I realize like, wow, like we uh we need to re-educate society.

Um, but one thing that gives me hope is that we did at Funstretch study another technological wave that wiped out at least 20% of the labor force um in in the 20th century which was frozen foods.

So, what many people don't realize is that when **Charles Birdsy** (Charles Birdseye: 现代冷冻食品工业的创始人) uh created Flash Frozen, which by the way was a venturebacked by **Goldman Sachs** (Goldman Sachs: 一家全球领先的投资银行和金融服务公司), it was a VC backed um company.

I love this.

And by the way, his name is Bird's Eye cuz he was an ornithologist.

He actually was studying birds.

Um, but he found that the **Anoui tribe** (Inuit: 因纽特人，北极地区的原住民) in Alaska had kept their fish super fresh and because they were putting it in a frozen saltwater solution that flash froze the fish.

If we look at the labor tables from the 20s, 40% of the US labor force was employed on farms.

It was literally we spent most of the economy was people working on farms and most of the service sector that was defined back then were household servants, people working for someone else.

food was over 25% of the wallet prior to frozen foods being uh widely, you know, mass market because most spoiled on the way to the supply chain and so grocery aisles were mostly fresh and what they what was frozen back then had freezer burn.

It was terrible.

So um flash frozen allowed suddenly the cost of food to drop dramatically because you had less spoilage and the number of people working on a farm today is down to what 2% of the US workforce.

So flash frozen was the really the key innovation that brought down the cost of food from 20% of the wallet to what is it five or 6% today and reducing farming labor from more than I think it was 40% of the peak down to two an economist in 1920 okay let's just pretend on CNBC in 1920 there is none but let's say there was a CNBC in 1920 and these economists were saying frozen food if it comes along and it's going to wipe out 95% of all farmers this is going to wipe out the US economy the US economy can't survive uh frozen food and instead it freed up time right and it created it allowed people to be repurposed and it created a completely new labor force so I so Scott to your point I think that there is an adverse outcome but then when I look at past episodes of huge labor disruption.

It's actually had positive outcomes.

every Okay, that was a lot of detail and storytelling to make the argument that AI will open up new opportunities and jobs.

But as you can tell from this pod, I mean, I really don't think it's just me.

You're getting a glimpse of him making a little bit of just a a sort of pivot in terms of his perception of how he's thinking about the labor market, how he's thinking about AI.

Obviously, a lot of data backs his research, which is fair, but it it starts to get very confusing when you make the argument that 90% of the stocks overvalued, 20% draw down, 40% upside after that draw down, and that's how the story plays out.

Like, it's just it feels like Tom is confused, and that's okay.

He's a human being.

He's, you know, he's allowed to be confused.

But nonetheless, it does feel like he's a bit confused here.

Let's bring someone on to the show that uh well, they've not been that confused about the story around semiconductor stocks.

Let's see if Broadcom or Oracle ended up potentially making him a bit more confused today.

Ladies and gentlemen, Jose, what's up?

>> What's going on, brother?

How's everything?

>> Everything's good.

We had a little bit of a crappy day today and I think uh you can definitely help us give some analysis on what's going on.

I think we'll start off with Oracle before we get to Broadcom.

We didn't talk since the earnings.

Uh Neil and I streamed this one together.

Which by the way, I kept calling him Jose.

I don't know if you you kept saying Jose and I didn't notice and like he would just look at me and I'd be like, "Why aren't you answering me?"

I'd be like, "Oh [ __ ] I just called you Jose."

So yeah, he's European Jose now.

But um

>> it's because you you you wish I was there.

Unfortunately, I was not unfortunately, but I I I had a had to take the little one to a few things that day.

What was your perspective on Oracle earnings and with that any thoughts on this Oracle opening data center delay that we got today?

What was your perspective on Oracle earnings and with that any thoughts on this Oracle opening data center delay that we got today?

What was your perspective on Oracle earnings and with that any thoughts on this Oracle opening data center delay that we got today?

>> Uh I I mean I thought Oracle earnings was a great insight of the AI market.

I mean they mentioned that this quarter alone they grew RPO by I forget the actual number but it was north of 60 billion just in one quarter.

Um and it wasn't just on OpenAI, right?

I think that's one thing that they wanted to do was kind of escape the OpenAI um bare thesis that was once a bold thesis now it's like a bare thesis.

It's funny how the market reacts.

Um but they mentioned that this RPO was mainly because of Nvidia, Meta and other names.

Uh so they they at least try to diversify and then within that RPO it doesn't have much changes this fiscal year which ends in two quarter but it did add an extra $4 billion to next fiscal year in their OCI and $4 billion I think last time they projected around 34 billion so 4 billion is about a 15% growth on guidance just on OCI which I thought is is impressive right just um just in in in the matter of a quarter uh because of that right?

They had to increase capex by 15 billion.

um capex I I think was 50 billion now um with the new guided that's numbers that like it's it's matching to the big boys right it's matching not as not necessarily to it but um it's getting up there right I think meta was like uh eight seven I forget the actual numbers I know most of them were in the north but Oracle is getting good is getting up there that I think was very bullish for AI chip companies in general because it means they're going to spend an extra $15 billion within the next two quarters because they just finished this quarter.

So this is for this extra 15 billion is for this fiscal year.

So an extra two quarters where they're going to spend an extra 15 billion than they expected.

Um so overall for the ecosystem and I'll throw it back to you just in case you have any questions but to the overall ecosystem I thought it was bullish in forms of AI compute demand and it was bullish for for the infrastructure players per se and then it was bullish for the semiconductor ecosystem because they're going to be spending more money in the um in that AI infrastructure itself.

Do you think then the decline of the stock from 220 to 189 had more to do with the debt that they said they're probably going to be issuing a lot more for this buildup?

>> Yeah, I I I think so.

I mean, I think for them, um I when you hear this crazy 500, I think sometimes the market just doesn't listen.

um they have this 523 billion RPO and most people are expecting it to be this year or next year.

They actually gave a perfect picture of it's actually like 16 billion.

I forget the actual number.

So don't don't excuse me for not being correct, but it's like something around 16 billion this fiscal year, 34 billion next fiscal year and then 64 and it continues to grow from there.

So, I think the market just wants a lot of that RPL to be more frontr run, but Oracle can't do that.

It's all about data center being built up and they're just building that data center to help the revenue.

So, I think the market is just hoping for that and yes, the debt obviously plays a big role.

Um, but I think **Larry** (Larry Ellison: 甲骨文公司创始人) uh counter back if I was reading the transcript and and he mentions that there's a lot of analysts that believe Oracle needs to raise over a hundred billion dollars to fund this and they said we don't need a hundred billion nowhere near that to fund uh to to raise to be able to fund it.

Obviously, they're probably going to raise maybe like 50 to 60.

Um but he believes that a lot of these analysts are expect are are are thinking they have to raise too much money.

Um and that could be an upside in the future.

Okay, Broadcom.

Uh, I think the reason they're down today, well, I'm curious what why do you think they're down?

And do you think it's because of uh the lack of OpenAI revenue coming in 2026, which is kind of consistent with Nvidia that says we don't really have a definitive agreement.

Why do you think Broadcom took the hit today?

>> Yeah, to be honest, I I I actually posted yesterday because I don't know.

I feel like Broadcom's becoming a very popular stock.

Uh, I honestly did not listen to the Aries call just yet.

Um I so I haven't have too much information to add there.

Um I wouldn't think OpenAI um would be much of a fear.

Um because I I they did announce **Anthropic** (Anthropic: 一家人工智能安全与研究公司) is a big customer now with the **TPUs** (Tensor Processing Unit: 谷歌开发的一种用于机器学习的专用集成电路) and and Brockcom and Google.

Uh so that's that.

Then obviously Google is still doing pretty well.

I think now the fear is I don't know.

I I I don't think I can give you any any any added value here on Brockham right now as I haven't read the earnings.

Um but it's just I I mean if you just look at that year to date year to date the stock has done amazing.

It could just be as simple as maybe it just ran up too much on excitement.

Um and and we're just pulling back a little bit but I think Broadcom investors are still pretty pretty happy for the year.

>> I think that's probably the best reason.

I mean this thing was up 80% year to date and they said a couple things on the call that whenever you get a chance to listen to Yeah.

not the most exciting, not bearish, but definitely not perfect.

And when you're trading at the multiple they were trading at, you got to be perfect.

Like Nvidia, they had to be perfect.

They were perfect.

Market still didn't care.

Broadcom wasn't perfect.

Obviously, the market's going to care to the downside.

So, yeah.

I I I wonder, right, what is this company worth now?

Two trillion.

Is it two trillion?

>> 1.7.

And then you have Nvidia worth at what?

4 point.

What is Nvidia at right now?

If you

>> 4.2

>> Yeah.

>> Right.

They mentioned a backlog of about 72 billion they mentioned right I think it was 72 and then if you look at Nvidia they have a backlog of not 500 billion a little bit less than that probably still about 400 billion remaining um so I guess just by looking at the backlog you can say am I really needing to value this at should I value this maybe a little bit cheaper than it is right now just because the RPO multiples aren't the same um on those uh but but yeah I think overall I mean the demand for ASIC is still going to continue to grow.

Uh there's a lot of they have that big customer with Google and they're doing really well with it.

There's actually even rumors that um they might have even won some Amazon as well.

I don't know if that was confirmed or denied during the earnings call, but um the compute is still extremely high.

I mean, we've seen so many partnerships with OpenAI and Enthropic just in the past five days that show that AI compute is not slowing down.

More importantly, Chris became extremely bullish on on on AI, it seems.

That's why we freaking fell today.

Freaking Chris buying Nvidia.

It's should have known.

Should have known that was the No.

So, what is your general sentiment on AI right now, Jose?

Because um Nvidia's multiple continues to look ridiculous.

Uh Broadcom was expensive, came back to reality a little bit.

Oracle probably never deserved to be at 350.

It's at 189.

That's not crazy.

Still 15% year to date.

I mean, is there been the Neo clouds are obviously are getting hit.

Uh, OpenAI is intimately tied with **CoreWeave** (CoreWeave: 一家提供高性能GPU云基础设施的公司), which is why I think the headline with with Oracle and OpenAI took CoreWeave down.

But has anything changed for you?

It's kind of just the market's tired of this trade and that's why we're consolidated.

>> Yeah, I mean I think for me it's it's a buy the dip moments, right?

I think the AI ecosystem, the AI story is still much intact.

I I understand why the market is scared, right?

I mean I don't think most people follow this market on a daily basis like like me and and able to have like these diamond hands that I have.

Um and I mean on the optics look scary.

It's like we don't know where AI is going to go.

You hear about delays.

Does that delay is it is the delay something talking about the financing which we keep hearing about is is this financial bubble is it about demand?

So when you hear just the story like that it just creates a whole fear of throughout the whole ecosystem.

Um but I think I mean for me I I I definitely am not worried about anything from demand uh from financing uh Nvidia's especially on Nvidia's part.

I mean the customers that are buying from there it's not just Oracle right it's your Microsoft it's your Google is your Amazon and all of them are saying the exact same thing about a month ago that demand is still growing they still want to continue to grow their market share.

um Anthropic and Google and and Open AI this week announced various partnerships with enterprise companies.

I think open AAI announced something with Disney which can be seen like AI slot coming to the overall platform but more kids are going to be just using AI in general.

More parents are going to be using more AI.

I do also think that's a uh it creates a domino effect for other creative enterprises to say look we can't be left behind.

If Disney is doing this with them, why shouldn't we should also go partner up with maybe not open AI but with some AI company because we need to have that same type of technology.

Open AI also partner up with anthropic partner up with **Accenture** (Accenture: 一家全球领先的专业服务公司) uh I think a few days ago.

So enterprise are the people that have the big money.

They're the ones that have those uh people on on computers eight hours a day.

Comput is going to continue to grow.

Um you saw **El Salvador** (El Salvador: 萨尔瓦多共和国，中美洲国家) also kind of announced that they're going to be using uh **Grock** (Grock: 一种人工智能模型).

More countries uh I I know El Salvador has been very popular lately and more countries are going to be like if El Salvador is doing this maybe we should also think about doing this because um we can't be left behind.

So I think the demand for AI compute continues to grow.

So I'm not worried at all about this AI sell-off.

>> Yeah.

Well, I agree with that as well.

If the demand for AI compute continues to grow and what we've seen from every earnings, which is undeniable, you you can hate the multiple some of these stocks were trading at like a Broadcom, you cannot deny they had a great quarter.

Same thing with Nvidia.

Nvidia's multiple a little bit better than Broadcom's.

Same thing with Oracle.

Maybe their multiple and their debt load not exciting, but they put up a good quarter.

Uh so I I think that is a fair point that S&P is up basically 20% this year.

stocks in including the most important companies on the planet have put up like all the hyperscalers all the max put up very strong earnings but also it's like all right from the liberation day lows forget year to date a lot of these I mean Nvidia is a double from the April lows you know at at 180 175 so it's like it's like as much as we think it probably should get a higher multiple if we're assuming that growth rate continues if it's not it's also like all right it is what it is um I'm I'm curious about your thoughts on this as well so China's opening up for Nvidia.

Nvidia said, "We're happy that Trump is allowing us to sell."

Uh, some articles came out this morning that said Chinese firms are basically begging the **CCP** (Chinese Communist Party: 中国共产党) to allow them to to buy some Nvidia chips.

Do you think this is a tailwind or do you think this ends up being a nothing burger for them?

>> Um, I I think for Vidia, it's definitely a tailwind.

I mean, uh, **Jensen** (Jensen Huang: 英伟达CEO) has mentioned in the past that this is a 50 billion opportunity for them um, per year and not it's not just about the revenue, it's also just about the overall ecosystem.

uh we've heard that a lot of AI developers just actually ends up coming from from China.

So you want to make sure that your developer ecosystem continues to grow.

Um I I I I think in forms of numbers is a tailwind.

In forms of the market though it becomes a headwind in the future because now once it gets approved um and once we start to sell start this we start to sell selling over here we're going to start to see reports uh potentially China um potentially United States is going to cancel this deal.

Potentially China is not going to take them and it's going to create this volatility in the overall market.

Um, so I do think it's a tailwind, but I think it's a tailwind that's going to come with the double-edged sword that uh the media and other players uh I don't want to seem like a conspiracy theorist, but it's just it it's headlines get clicks and scary headlines get the best clicks.

So if you ever have just like with this Oracle thing, um you're going to see my next question.

What was what was your perspective on that when it came out?

>> on which one?

My apologies.

>> on on OpenAI delaying a data center theoretically with Oracle.

Yeah, I I I mean and I I thought the market was funny, right?

If Oracle code delays uh well, we saw that it was false or or they they dispute the fact.

Um but first I understand why the overall ecosystem fails like again you I think you mentioned it.

It's like it it's either means someone has slow demand, someone's not able to finance and it just trickles down to the rest of the ecosystem.

If Oracle though, in my opinion, the way I see, if Oracle can't meet the compute because they slowed down on some form of data center, it has to trickle down to someone else.

Open AAI, if they really need the compute, they're not going to wait.

They're going to say, "Oh, maybe I have to go buy it from Nebus, from Corwe, from iron, from someone else, from or from Amazon, right?"

So, I thought the market was pretty funny because if it's a slowdown for Oracle, that demand needs to go somewhere else.

And that's why I found that the whole market kind of going down.

Um, sometimes it's it's it's it's funny.

Uh, I don't think it's true.

Um, and I don't think it's anything with compute.

Oracle just had earnings two days ago or three days ago.

And they mentioned that their guidance for OCI for 2027, fiscal year 2027s, which is about two quarters ahead of calendar year is up.

Why would they have done that if they expect some form of delays?

Why would they kind of create some form of fear if they had some form of delays?

But instead they actually upped their guidance for next year.

Um so I mean the data was right there I if in my opinion

>> and and capex was still high.

Yeah.

>> Yeah.

And and they increased their capex guidance as well by 15 billion.

I I think the market's just tired of the AI trade right now, right?

Like and I think I think you're probably feeling it as well.

I think a lot of us are feeling it that you know the market's given a lot of these names a multiple.

A name like a Nebius.

A Corby is still up but double from IPO.

IPOed at 40.

It's basically at 80, right?

Neb is up like 2.2x or something.

So yeah, these names are not at alltime highs, but they've definitely made a move.

It is now their job to keep executing to prove they deserve their multiple.

But are do you think the market's just a little tired of the AI trade and maybe that's where, you know, you get some consolidation towards the end of the year.

>> I I I think more fearful than anything, it's just, hey, we just had this massive drop.

A lot of these stocks I I I think right um Nebia's dropped from like 120.

core we've dropped from like 140s.

A lot of these were are down probably still 30 or 40% and it's like during this time frame everybody's scarce like uh any any little noise it's like they they they still have PTSD from from just a month ago and it creates that negative sentiment.

Um and I mean Cororee unfortunately didn't do much help to it when they reported earnings.

They mentioned that they had some delays on the data center side but it was more of like just the construction aspect.

It had nothing to do with demand.

Uh but the market likes to run with it.

So unfortunately I think we need a a quarter where like everybody is perfect where everybody that reports says look this is good good good.

Um to kind of maybe turn that sentiment but I I I think it's yeah I think everybody's tired.

I think everybody has that fear of oh I dropped 40%.

It's going to drop another 20% from here.

Um, let me get off before uh emotions are high on the fearful side for AI investors right now except for me.

>> Okay, so my last question for you and guys Jose's YouTube and Twitter.

I'll put in the chat.

You can follow him there.

Uh, yeah, I guess simple last question.

Why why don't you have any fear?

>> Well, I mean, it's I think mainly it's just because of of my workflow.

I continue to use more AI on a daily basis.

I've c even I I come from an engineering background, so I do have some tech stuff, but I've created like small apps that I use on my computer now that I I wasn't able to do back then.

Even with like the knowledge that I had, but with AI, I'm able to do this stuff in like minutes um in in hours, I'm able to create some crazy things.

I I talked to friends.

One of my friends is actually trying to um start a business and I I went over his house a few a few days ago and he he showed me how like he's incorporating AI so much into this and it's like it just blows my mind how much how much AI is growing in real use cases.

Uh I sometimes do some form of freelance for for various companies and again the amount of AI work I see is wild.

So to me it's just real data um from kind of boots in the ground that show me that this AI story is not going anywhere.

And then with a lot of these companies right some of them might have some crazy valuations but Nvidia at a **forward PE ratio** (远期市盈率: Forward Price-to-Earnings Ratio: 基于未来预期收益计算的市盈率) of like what maybe mid20s right now if that uh the Amazon um Meta like they're not that crazy.

They've turned a bit around but valuations aren't scaring me for for a lot of these players.

And then for my yellow, not my yellow players, but for like my ones that I think are going to are going to make big money for me like Cororey, like the demand for compute continues to grow and the and these are players that are building that massive AI infrastructure.

Um, so yeah, it's just I think boots on the ground data.

I think that's fair and I think anecdotally a lot of people that are actually using AI every day and sort of seeing the the technical aspects of not only model improvement but how the application of those models are leading to actual productivity gains in their own life.

That's a big deal.

And then when we get the evidence around, you know, the earnings growth and the data center growth, it's hard to fundamentally deny it as like a a fad as some would say.

Um, okay.

Actual last question for you, Jose.

We're we're getting towards the end of 2025.

So, if you had to give me three of your top AI convictions for next year, what would those three be at this point?

>> Uh, Nvidia for sure, especially at these levels.

This one, even though a $4 trillion company, I could see this being a massive player.

So, Nvidia for sure.

Um, number two, Core Weef.

I believe Core is and and it's crazy to think I always hate when I say Cororeef because many people think I hate the other players.

I don't.

I think the whole Neocloud ecosystem is going to do great.

I just prefer core the most because they have the biggest AI infrastructure out there over about 600 of active power.

They're like best buddies with with **Jensen** (Jensen Huang: 英伟达CEO) and Nvidia.

Uh so so Cororeef is my second.

Um and I have to stay with the third.

For me the third I guess I'm going to go with AMD.

AMD is that one that's like a $200 billion uh what's I forget the market cap probably like 350 right now.

And if they do reach those earnings per share potential of $20 within the next two to five years, that's a trillion dollar company easily.

Uh so those are my three favorite plays.

It doesn't mean I hate other semiconductor stocks.

And I'm sorry to the viewers if I didn't mention yours.

Don't hate me.

Uh but in general, I think AI chip companies.

I think the NeoCloud players are still going to do very very well um in 2026.

>> All right.

I love it.

Jose, thank you for joining on this Friday.

Have a great weekend.

We'll talk to you next week.

>> Thank you.

You too, brother.

>> Later.

All right, folks.

That's what we got from Mr. Jose.

I do not think Steve is joining today.

So, yeah, we're probably not going to be hearing from him.

I don't even think, quite frankly, Steve even saw what happened to the market today.

He's been texting me pictures of his lunch all day.

So, the other thing he texted me about was this freaking **OC** (Oracle Cloud) news for **XRP** (瑞波币: 一种加密货币) because for some reason he still loves XRP even though he doesn't buy any it any any more XRP.

But yeah, outside of that, I don't think he cared too much about this decline today on Oracle.

Again, I I think it's just it was a little bit weird because last night alltime highs on the S&P, then this morning we're doing okay and then boom, Broadcom bottom falls out.

Open AI news comes out with Oracle.

Oracle denies it.

market recovers and the market says, "You know what?

We actually don't believe it."

And you end up just getting a 1% down on the S&P.

Not to mention, it's a Friday.

So, there's tons of options that expired today and probably had a bunch of different stuff going on with market makers.

And that's pretty much what we had.

He was on with **Tanner** (Tanner: 某人名).

Okay.

He was probably on with Tanner then during the the uh the middle of the day.

He probably did.

He said like his his his company like ordered lunch for everybody, but he used it as an excuse to go get his own lunch because they like canceled a bunch of meetings because they all ordered lunch at his office and he went straight to **Paty's** (Paty's: 某餐厅名) to uh to get his meatballs.

I'm not going to call Steve.

I already sent him the link.

I already texted him.

I said if you're free, you can hop on.

If not, it's okay.

But he's probably stuck in those meetings that the company canceled during lunch.

and now he's there after hours.

So, if he's free, he'll let me know.

If not, then uh he'll be live tonight.

Also, that was we were live together for two hours yesterday.

So, Tanner called him for like two minutes.

Okay, that makes sense.

Usually Fridays tend to be a little busy for him.

Um yeah, I mean, Mr. Tick, no crying at the casino.

There is no crying in the casino.

Again, I think the the thing that really surprised me today had nothing to do with the markets.

It was this Tom Lee podcast.

It was this freaking Tom Lee podcast, dude.

I mean, you guys can listen to your listen to it again.

I just I I this was the because look, what's his name?

Uh Scott Galloway had Osweeks ago and he was straight up bearish.

He was basically like sell everything.

It's over.

Got a lot of people scared and then we went down.

We recovered.

hit alltime highs yesterday.

Then Tom Leehop's on the same podcast and he's like, "Yeah, I think we're going to be okay next year, but we're going to have to go through some real pain before we're okay."

And I'm like, "Okay, what does that mean?

20% down."

I'm like, "Okay, so where's the target?

40% up after 20%."

I'm like, "So, what what do we what what are you saying?"

Like, genuinely, what is the argument here?

It's it's just confusing to me because I don't know how realistic it is for the thesis play.

If that look, if we have a 20% draw down, small caps obliterated, financials obliterated, uh, crypto obliterated.

Like all of his picks that he's excited about next year, they're gone.

If you have a 20% draw down, everything's gone if you have a 20%.

I mean, Verizon might be okay.

But Tom also thinks the Supreme Court decision on tariffs will cause disruption next year.

I think he said that yesterday on CNBC.

So, yes, I mean, that's another that and the freaking shutdown, those two coming together could be an issue.

And then unfortunately those poly market odds about um the shutdown.

By the way, Trump is not live right now.

I think he might be live later, but he's not live right now.

The shutdown odds they increased today.

And yeah, right now they're at 40%.

And that's not going to be good.

That's not something that we need to see.

The weed stocks were safe, though.

weed stocks went higher today, dude.

Like, we cannot have a shutdown going into because because remember the first shutdown, the one that we had like a month ago, I mean, economists are saying 2% of GDP got wiped out with that.

We we cannot afford with the debt we have and the deficit we have to have another 2% on GDP get hit.

Also, we have the jobs data next week, so that's going to be important.

Now the jobs data being see here's the issue now.

Let's say jobs data is bad.

There's no more rate cut catalyst because the rate cut happened.

So it's not like bad news is good news anymore.

Bad news might just be bad news.

Now people might say oh we'll get another rate cut in January.

Maybe maybe that's where the narrative is like okay more really bad jobs data means even more rate cuts.

Right now January is pricing in a 25% chance of a rate cut.

But it might more so just be bad news is bad news which is okay.

We already know the labor market's bad.

So, it's like you can't avoid that data being bad and the Fed probably knows it as well, too, which is why they cut this week, but that could end up being the next catalyst.

It just will take a month before we see if we get that rate up.

Yeah.

Trump said Kevin Worsh has moved to the top of his list, though Kevin Hassid also remains in contention.

I mean, Worsh would be the safe pick because the street loves him.

Uh Wars actually interviewed **Alex Karp** (Alex Karp: Palantir CEO) if you guys remember that was back in January of 2022.

But um Hassid would be loyal.

So I think Trump's got to figure out who does he want here?

Someone who he knows is going to cut rates or someone who he knows is dovish enough to cut rates but might not explicitly end up cutting those rates.

Yeah.

And then going back to this Tom Lee 20% draw down thesis, we got the midterms coming up.

uh we get a 20% draw down and if that's anywhere near and not get a 40% recovery as Tom Lee is saying going into the end of 2026 November which is when we have midterms I don't know if people lose 20% of their retirement accounts they might not be the happiest going into midterms.

So there are some I think broader concerns on that entire thesis, but I'm curious about what other analysts have to say, right?

Like so this is funds perspective.

I'm really curious about like what does **Morgan Stanley** (Morgan Stanley: 一家全球领先的金融服务公司) say?

**Mike Wilson** (Mike Wilson: 摩根士丹利首席美国股票策略师) has been pretty damn bullish and I I think his take might be a bit different in terms of how it plays out, but still on the same track record of maintaining a bullish view.

And again, my perspective is very simple.

Fed easing, AI trade intact.

You put those two together, you get a bull market.

I don't know if it's 10% up, 20% up, 14% up, you know, 11% up, which is still a great return after three years of 20%.

But I don't see those two collapsing next year, especially with the new fetch area.

Now, they're trying to collapse the AI trade with every single headline every other day.

Uh, and we had that taste that today with Oracle and OpenAI.

But if those two narratives stay intact, then I think we're going to be all right.

Uh Tom's hurting because of ETH.

Well, that's another thing, dude.

That's I think that's the other, you know, I I said this when BMR was getting hit pretty bad in November.

He got the timing of it dead wrong.

Regardless of what you want to say about the thesis, which I think the thesis is very real around **stablecoins** (稳定币: Stablecoins: 价值与美元等稳定资产挂钩的加密货币), **tokenization** (Tokenization: 将资产转化为区块链上可交易的数字代币), that stuff's only growing and growing and growing.

The timing of it was dead wrong.

So then the question becomes, okay, if we're going to have a meltdown as he's again, I'm not saying we're getting a 20% draw down.