Oh, this story is just getting crazy. It turns out that Donald Trump, who is relaxing for Jensen, has a stake in Nvidia worth $62 million, or at least that's what it was worth in late August when they last reported it, which means we're down a bit. The stock market is selling off, and it's all because of Jensen Swank. I mean Jensen Huang. In any case, the stock market is selling off. Part of this is because people believe Jensen was subjected to a psychopathic attack by the New York Times. We'll talk about that towards the end of this video. More importantly, what is potentially the most optimistic research firm on AI chips is suddenly lowering its spending projections by $1.6 trillion through 2030. In
other words, the additional $1.6 trillion they predicted will disappear. Zero, baby. This is crazy, and all of this is happening right after the Jensen debacle, which makes you wonder how many people knew we would have insiders starting to rally against Jensen. I mean, this bullish AI research firm is currently discounting Nvidia by up to 60%, much more than other Wall Street firms are doing. I wonder how many put options they bought on Nvidia. And then you literally have Scott Bessent , who started an anti-Nvidia campaign against Jensen Huang, claiming that only 1% of Nvidia chips sold to Singapore actually stayed in Singapore, which of course begs the question, why didn't the government actually step in and do something about it
Probably because the government secretly, uh, or rather openly, just kept selling. This is good for GDP. But anyway, to be fair, Singapore had Nvidia sales in 2024 And now Nvidia is rethinking where those sales will be recorded, moving them from the companies' headquarters, not necessarily where the chips come from, perhaps to hide where those chips are actually going This reiterates why people like Bassin on Fox News say: Nvidia needs a shakeup. The next CEO should be from outside the company. I can't shake the idea that Scott Bassin fears for his job, that if for some reason Donald Trump supports Jensen, and the finger is pointed at Latnik or Bassin, they will find themselves on the receiving end of the fire. So Bassin should act like, no
, no, we tried to stop Nvidia. Nvidia is bad. This is quite interesting, especially after about an hour and a half ago Donald Trump came out and said, "I certainly wouldn't mind Jensen coming back And that Jensen learned a huge lesson. This comes at the same time as staff are discussing a demonstration at the Santa Clara campus. We'll see if there's a demonstration at their headquarters. 175,000 people signed a petition asking for Jensen to be brought back and supporting him. They actually had such huge traffic, or website traffic, that the website started to become unstable. Uh, to be fair, there are some
articles that are coming out that are suggesting that, you know, Jensen was also trying to limit the open checkbook quotes, that the days of Nvidia's open checkbook funding, uh, are coming to an end anyway, so, you know, people should stop relying on Nvidia's cyclical funding. So, you know, there's some bearish news on the AI cycle, but I think the biggest bearish news isn't Jensen telling Neo Clouds, "You're underpricing." He told them that Elon Musk was leasing Nvidia chips to customers at better prices than they could afford themselves. I don't even need you guys. I could just give all that power to Elon. "Elon wants it." Which then makes you wonder, isn't
Elon guaranteeing the whole AI boom, right? But let's put that aside for a moment and understand what just happened with this semi-analysis fall, because everyone on Fin Twit is panicking about what happened to semi-analysis. Semi-analysis, in essence, significantly reduces its forecast. This is spreading everywhere now. Uh, that's what people are looking at. Semi- analysis capital cost estimates. They are actually leaving the 2026 forecast unchanged to save face. They reduce capital expenditure estimates for 2027 from 770 billion to 627 billion, a 19% decrease. From 1.05 trillion by 2028, it will decrease to 758 billion, which is 28% less. Then the 1.35 trillion by 2029 is reduced by only 796 billion. And 1.53 trillion
by 2030 is reduced by 40% to 923. It is worth noting that the second derivative here drops sharply. This increase in capital spending between 2028 and 2029 is only 5%, according to this analysis. That's the problem, okay? Semi- industry analysts knew the vast majority of this nonsense before they suddenly decided to change their minds. Just 11 days ago they repeated their numbers about how optimistic they are and everyone else is not optimistic either. So let's be realistic. What has really changed in the last 11 days? Because that's what matters here. If there had been any major changes in the last 11 days, perhaps that would have been fair. I mean, Jensen really did resign. We had demographic
cleansing. Are these the things that have changed? Well, let's analyze this a little. As for what happened to the physical hardware, nothing has changed. Literally nothing has changed. You have companies around the world still ordering, placing orders, and growing order backlogs at either the same or faster rates than what they were 11 days ago. So either the semi-analysis turns into a sunflower child, and they're just scared because the stock market is going down, and they try to tell everyone, yeah, you see, our analysis actually justifies why the stock market is going down. Or they're just being dishonest about how good their
models are, because obviously if you need to adjust your models by $1.5 trillion or you know you need to cut $1.6 trillion from your models in 11 days, your models probably are n't that good. Yes, the midterm elections took place, but the midterm elections are only expected to delay data center construction by about 5.8 gigawatts between 2027 and 2028. If you put $ 50, $60 billion on 5.8 gigawatts, then you're talking about delaying costs by perhaps $340 billion from 2027 to 2028, that is, actually postponing it. This does not mean suddenly eradicating $1.6 trillion. So I think what happened was that the semi-analysts woke up from a very bad hangover or with a very bad
hangover. I think they realized they were drunk on their numbers and you had this ridiculous version of a Tom Lee-style half-bowl with chips and then they woke up and realized, wow, we were drunk. We should reconsider this because the stock market is falling right now, and we don't want to seem drunk while the stock market is falling. In other words, I honestly think they are simply adjusting their numbers downward to give the impression that they are not drunk and that they are simply watching the market, as Wall Street analysts do all the time How do we know this? Because when you look at the details of what they're saying, they literally bump their fancy power figures per chip from 12,200 watts to 16,000
, um, watts per connector, which suddenly can't be plugged in anywhere. They just cut it by about 50 %. They say, "You know what?" We are going to reduce the amount of connected energy by 50%. Then they assume that memory prices will drop by 18-30%, which, to be honest, most people on Wall Street were already predicting that memory prices would slow down long before this election. And then they decide to cut into Nvidia's future margins by assuming that OpenAI, Thropic, and Amazon will just magically use only cheaper ASIC chips of their own production, which is real competition, but Nvidia's GPUs will effectively be thrown in the trash. That's what I'm saying.I'm obviously exaggerating a bit
, but there's only one way to say that in 11 days, Nvidia prices will drop 44%, while ASIC manufacturers like Broadcom will only drop 6 In my opinion, the only way to reconcile this is through fear, mood, and credit. That's all. The VIX jumped to 28 on November 5th. Then we got a ratings drop after Jensen cheated, uh, you know, on November 11th, 12th, and 13th, we got a few ratings drops. We saw CDS growth and a peak in October for, uh, you know, funding neoclouds like Cloudflare, uh, you know, even meta CDS went up. And suddenly the stock market started to really fall due to Donald Trump not making a deal with Iran and the Federal Reserve becoming more hawkish Combining all of this with the Democratic defeat leads to a
stock market crash. And I think the semi-analysts are afraid of looking foolish, so they cut back on their silly analyses and try to look a little more sober to justify the stock market crash. But I really want you to think about it for a moment. Think about all the nonsense the stock market has had to digest. The Fed is holding the stock in October because it thinks Donald Trump will make a deal with Iran. Donald Trump is not making a deal with Iran. The Fed must become belligerent. This is one thing. No deal with Iran, belligerent Fed, inflation. All this is bad. Inflation rates are rising rapidly. The labor market is still strong. None of this is a
recession. This is simply a sign that we have an energy crisis. Good? On top of that, you get a "sweep" by Democrats, which, to be fair, will probably delay the construction of some data centers for years to come. Not as much as the semi-analysts are saying now, but it adds up.So now you have Iran, uh, no deal. You have the Fed raising rates. You have delays in building data centers due to the "sweeping" by Democrats. Add to that the Jensen cheating scandal. Holy God, dude. You have a lot of pain that this market now has to absorb. But when you actually look at what the semi-analysts are saying, their changes in numbers are crazy bags.
They literally remove $1.6 trillion in spending. And how do they do it? Well, they're assuming they're going to have to take big cuts, big cuts to Nvidia's revenue in 2027, '28, '29, 2030 They are discounting this entire stack by 32%. Morgan Stanley literally has 1.7 times the semi-analysis numbers for Nvidia. And in their analysis, Nvidia gets the vast majority of its hit or miss. They literally cut Nvidia's data center revenue by 66% by 2030. I swear these people love something like that when it comes to Nvidia. But in any case, two- thirds of this is due to Nvidia - what they call wasted silicon area In other words, suddenly we're going to see a massive reduction in pricing power
because we're going to reduce the silicon area Now it's a little more complicated. I'll give a very simple analogy here, but I mostly like to use the kitchen analogy. Imagine a kitchen as the kitchen in your home. The more square footage you have, the more cooks you can have. Chefs are like transistors doing the work of artificial intelligence, okay? The bigger your pantry, the bigger your refrigerator door, the more throughput you have to get the job done. The door is a kind of bottleneck. And then how much energy you get from your refrigerators or stoves, for example, gas
, means how much you can actually cook, right? The AI industry as a whole right now, even the silicon analysis says, or, uh, semi-analysis, semi-analysis says, "Oh, we're only going to reduce our projection of the square footage that we're going to get in kitchens by 15%. So everything is going to stay the same except we're going to slightly reduce the size of our kitchens by demographic cleansing or something by 15%. But we're going to assume that energy consumption is down by 50%, and we're going to align that so that we say Nvidia is going to take the biggest hit. That's not, in my opinion, evidence of anything that's happened in the last 11 days, other than the semi-analysis waking up and realizing, " Wow, Nvidia stock is going down."
How are we going to justify Nvidia stock going down? Why didn't our models tell people in advance that Nvidia stock was going to go down? Oh, uh, why don't we just blame Nvidia now? That's what you get. I mean, you look at it. on the chart we have here . Nvidia is giving up 40% of its markup. Okay, chips are suddenly going to face a massive power outage, and so we're going to lose $1.6 trillion in reduced costs between 2027, uh, and 2030. And suddenly Nvidia's chip revenue is going to fall by 66%. That's the bear story today. That's why the markets are red today. It's also relatively why even other companies are hurt. Semi's analysis shows that Broadcom will have to cut its prices by 6% and AMD will have to cut its prices by 10-15%.
So overall they're trying to be a little more reasonable in their numbers, but they're coming out of this crazy high state. Anyway, Semi's analysis says that up to 30% of their cut is due to lower memory prices. Now Bernstein is saying the opposite. They actually think that contract prices for high-speed memory will roughly double in 2027, and that memory prices will still be the main driver of semiconductor price increases But semiconductor analysis is now using the flip argument and saying, "No
, the cost of high-speed memory will go down." That will reduce margins, that premium that can be charged for chips like Nvidia's, which essentially package high-speed memory and resell it at big premiums because of the rising cost of memory. And that ASICs will take over the initials. Okay. All of this sounds very complicated. The main argument here is that a big research firm that people think of as a big bull on Wall Street is now saying, "Hey guys, we're going to be less drunk, uh," and part of the reason we're going to be less drunk is that we think Anthropic growth is slowing more than we thought before, which is really interesting because the pace "Anthropic's growth was already declining, uh, before we even
got to this point." This is where I get so confused as to why these people are suddenly changing everything. You see, 11 days ago we already knew that Anthropic's growth rate would not be the same as last year. The big growth came in the first and second quarters of this year, so the annual numbers will of course be lower. We knew this 11 days ago. We also knew that funding had become tighter 11 days ago. Credit default swaps peaked in October. We already knew these two things . The only thing that has changed is that we are seeing a delay in the commissioning of data centers of 5.8 gigawatts, you know, actually. And this is a delay, not a cancellation. We don't see any cancellations or delays from
Meta, Microsoft, or Google. They are all spending like crazy. For that matter, they spend and borrow more. So two of the three excuses they give for changing their numbers, we already knew when they repeated their optimistic numbers. The election took place, but it didn't have that much of an impact. So you basically have a company that says, "Hey guys, the stock market is going down, so we're just going to follow the market in our analysis." I think that's exactly what's happening here. And I mean, you know, don't sue me, bro. I'm not trying to blame them. This is my opinion. I could be wrong, but I honestly believe that this analyst firm got caught up in over-optimism, and now they're trying to justify why Nvidia
is down 34% from its peak, and here's how they do it, using mostly old information that the market already knew, and then trying to magically say, "Oh yeah, Nvidia will take the brunt of it." Whatever. I don't care. I have very little experience with Nvidia. I sold the vast majority of my Nvidia stock last year because I had to pay a big depreciation expense bill, and I made a huge profit on Nvidia, so I sold millions of dollars, and Nvidia makes it so that I have so few shares that it doesn't really matter. I'm not trying to promote Nvidia. But it raises questions about this whole idea that labs, like Microsoft's data, show that Claude is lagging behind Copilot.
Well, this is really good publicity for Microsoft, especially considering that Anthropic is starting the week of its IPO. You see, Claude for Microsoft is less than 1/10th of the number of weekly users of the paid version of Copilot - that's what Microsoft is trying to say. They are essentially trying to say that fewer people are using Claude and more are using Copilot. For some reason, I find this very hard to believe.
Uh, some of the analysis also shows that, you know, more policies like the Remote Access Security Act are now going to be passed, and, you know, the chances of that being passed are now increasing. It has stopped twice already. Every vote against was for Republicans. So yeah, is it possible that this will pass now because Democrats will control the Senate? Perhaps. But Trump will simply veto it. And it seems to me that they don't even consider it. So this is, you know, an attempt to limit the supply of chips, you know, to foreign countries, essentially. So I really disagree with that.
Regarding the psychological operation and the Anthropic IPO, that's interesting. There is now analysis that US intelligence, essentially, memos from Marco Rubio at the CIA, ousted Jensen Huang. But the whole reason this all became public was because, essentially, the argument is that the Jews who run the New York Times decided to coordinate this to get Jensen out of Nvidia. Now you might ask, well, well, what do Jews have to do with this? Why are we suddenly talking about Jews? Well, here's the thesis, and it's mostly conspiracy theory, I think, almost entirely. But in any case, Jack Clark of Anthropic was reportedly the one who
leaked the spy story about Jensen to the New York Times. But during Jensen's October tour, he publicly expressed support for Open Weight, essentially opposing Anthropic and OpenAI. People believe this threatens Israel's control over the entire AI industry, as they have people at the helm of Anthropic, OpenAI, and Palantir. This is very conspiratorial. Uh, I just want to address this. I don't think this was an Israeli psychological operation to eliminate Jensen because he supports Open Weight. Uh, but I want to address this because some people think this is actually happening. So, what's
next? Tuesday, November 17, Nvidia announced. It's tomorrow. Without Jensen. This is going to be massive, okay? I kind of expect that they will still raise rates, and they will still say that the backlog is growing wildly. We'll see . We'll look into this tomorrow. It will be interesting. November 25th is 10 days since Bessant's call for answers. Thursday, Anthropic IPO. If Anthropic collapses, the entire market will continue to fall. Okay, we have some predictions that we talked about in the live broadcasts for course participants. We hope you will join us at reinvestone.com. December 3 is the date Tim Scott tentatively wants Jensen Huang to testify before Congress. We don't know if it will be public or behind closed doors.
And then, on December 8-9, we will get the next Fed rate decision. So, how do I reconcile all of this personally? When I personally put it all together, I think SemiAnalysis is adjusting their numbers downward to save face and claim they knew the sell-off was coming. Complete nonsense. I don't trust these people. I look at their articles and then I leave drunk. I've said it before, I'll say it again. I don't look at their numbers. I trust what I see in companies' earnings, earnings forecasts, and forecasts from analysts who actually talk to those same companies. The SemiAnalysis numbers.I don't know if they're paid by Leopold or something. In my opinion, I don't like them because I
think they were too blindly optimistic, and now suddenly they're following the market down. I think it's dishonest and disingenuous, frankly, disingenuous. I think this lacks honesty. And I think it's honestly blaming Nvidia when they fall. I'm not here to say, "Oh, you know, you know, buy Nvidia drops." I don't care if Nvidia goes to zero. It doesn't matter to me. Do I think Nvidia will fall to zero? No. I actually think we are getting ready for a big bull run. I know, I know, people say, "Oh, Kevin, this is crazy." I know, I know. Let me read this to you, okay? Because I find it very interesting. So, by the way, I hate agreeing with institutions. I'm really against institutions, but I want to read this to you,
okay? So, here's what Morgan Stanley says, the AI capital spending cycle we view as intact. A resilient global economy and steady earnings growth, set to continue through 2027, provide a constructive framework with key risk catalysts behind. Risk assets have fallen in price over the past 6 months, and we enter 2027 with a generally positive risk stance. I agree. I don't like to agree with them, but I do. Here is another analysis. I think it was JP Morgan. We expect global growth to moderate somewhat in 2027, before returning to 2026 rates in 2028 as energy headwinds recede. The US remains the main engine of global growth There are clear challenges to the global economic outlook, mainly due to energy. However, we expect
these obstacles to disappear by the end of 2027. Continued inflation is an obvious risk that the Federal Reserve will have to deal with So, in my opinion, I know it sounds weird. The indicators are literally at 666 right now. I know. But it's simple . Stay away from margins, stay away from debt. Buy stocks, baby.