YouSaid · the spoken record

Remi Tetot

lines on the record: 77
first: 2024-03-03
most recent: 2024-03-03
sittings or episodes: 1
sources: podcast

Every line below is reproduced as it was said and linked to the record it came from. Nothing here is summarised or generated. Directory · Search · Corrections

“Know what? You put three things in that post, and I was like, you actually got a point. He's right. Like, we didn't have shit tons of IBO and we didn't have the fraud yet. Yeah, if you look at it as a bubble, yes. But, you know, like I said before, I use the business cycle as a framework. That's what I learned at GMI with Raul. Before COVID hit. We were in the longest. Economic cycle, basically, expansion, right? What happens is in 2018, 19, we were slowly, slowly digesting the excess of it. It was basically, the business cycle was playing out slowly. It was happening. And COVID happened. So it collapsed rapidly and bounced like even further on those crazy stimulus and stuff. Now, I do think that we haven't digested the previous alcohol yet properly because of COVID and because of the liquidity input and all the”
2024-03-03 · Forward Guidance · Remi Tetot on AI, Crypto, and Uranium · IDENTIFIED FROM THE TRANSCRIPT
“Going to you never know, but I feel like, you know, I feel like the hype is not that high. And it's only the high quality companies that are generating so much cash flow and growing so much that are earning. Like I feel like I don't see any, you know, it's a derogatory term, but shit co like a shitty company”
2024-03-03 · Forward Guidance · Remi Tetot on AI, Crypto, and Uranium · IDENTIFIED FROM THE TRANSCRIPT
“Internet was expensive now. It cost a fright, it was expensive and it was low. Now 20 years later, it cost a fraction of that for like crazy speeds. The beginning of that was those, the 2000 when they started to build those networks. Same will happen with AI. It's a structural technology improvement.”
2024-03-03 · Forward Guidance · Remi Tetot on AI, Crypto, and Uranium · IDENTIFIED FROM THE TRANSCRIPT
“Now it's the same thing with data with basically processing power. So all those services in the future, they're going to need a lot of processing power. And those processing power data centers are being built right now. That's the CapEx. That's the investment. That's happening now. For AI to explode, you need to have a burst because people are going to be able to buy all that power for really cheap. And then the end product is going to become very a lot cheaper for the end users. And eventually the user is going to end up winning because it becomes really cheap for them. That's the thing when they buy those data centers. They're going to make the product a lot cheaper for us to use a lot. And that's when you can really scale it. I think you have the price, which is really cheap, and you have the processing power. I mean, look at, you know, broadband like in 2000.”
2024-03-03 · Forward Guidance · Remi Tetot on AI, Crypto, and Uranium · IDENTIFIED FROM THE TRANSCRIPT
“It is of itself the cycle, that's for sure. You know, people oversold it. In 2000, when they wanted it, everyone wanted to build better network because everyone wanted internet. Everyone was buying a computer home, you know, all of that.”
2024-03-03 · Forward Guidance · Remi Tetot on AI, Crypto, and Uranium · IDENTIFIED FROM THE TRANSCRIPT
“So, in that future, it sounds like AI has a big place to role. So, is your view on, I guess, NVIDIA or AI, are you kind of a cyclical doubter? Do you think the cycle is going to turn? But the future is bright in 2030 and that much of the hype or maybe is kind of justified. And it just cycles gotten ahead of itself like it did in 2000? Or are you a cyclical and a secular doubter where you think, nope, it's overhyped kind of forever?”
2024-03-03 · Forward Guidance · Remi Tetot on AI, Crypto, and Uranium · IDENTIFIED FROM THE TRANSCRIPT
“Yeah, yeah, it's going to be those other way around You know, if you're skilled and you're good and you can give a human input, then that's going to be worth money.”
2024-03-03 · Forward Guidance · Remi Tetot on AI, Crypto, and Uranium · IDENTIFIED FROM THE TRANSCRIPT
“Now, after that, you know, it's like I said, it's all about the use case, like someone that works in a restaurant, is he going to need AI? Well, I don't know, unless you remove the waitress, for example, in McDowell through an AI machine, you know, for example. I also think that in 10 years, human input is going to be a luxury.”
2024-03-03 · Forward Guidance · Remi Tetot on AI, Crypto, and Uranium · IDENTIFIED FROM THE TRANSCRIPT
“Well, I think first making movies is going to change a lot, that's for sure. Basically all the fielders in movies, they're going to be made with AI now. You know, when you see a guy who's traveling and you see the plane in the sky and that's going to be AI generated, they're going to do that very quickly, stuff like that. All the fillers. Eventually, they're going to also master anywhere and, you know, filming people. And they're going to increase. So here's the thing. What are they going to do? They're going to improve the productions, but they're not going to make them cheaper because they're going to spend the money somewhere else anyway. So eventually, you know, maybe they're going to do more marketing or whatever. What it might give is the opportunity to skill people to produce cheaper good stuff. And that's for the video side. Well, customer support, obviously, for anything is going to be AI driven in 10 years, all of it, that's for sure.”
2024-03-03 · Forward Guidance · Remi Tetot on AI, Crypto, and Uranium · IDENTIFIED FROM THE TRANSCRIPT
“This week with the response he was giving you and stuff. So, I mean, this is actually very important because if it's going to be used at scale and they control what the content is being given to people, you basically kind of control people. So it's, you know, it's still dangerous. Now I'm waiting to see if ChatGPT fixed that in 4.5. I heard they say they were not going to be so woke, but in the end The problem I think they're going to have is if they use the data set from people that are woke, it's going to become woke because that's how it's getting trained eventually. You have all that risk of what data set you include in there and stuff.”
2024-03-03 · Forward Guidance · Remi Tetot on AI, Crypto, and Uranium · IDENTIFIED FROM THE TRANSCRIPT
“And now, the other thing is now because everybody runs after the same thing, now you have, so you have ChatGPT and you have Gemini co-pilot, the market's going to be quickly saturated with too many demand, too many of us, sorry. And if there is too many of her, people are not going to be able to choose because you give them too many choice. And then they're going to end up like moving away from it. And you also have the problem. We still have the problem of we've seen the problem with Gemini.”
2024-03-03 · Forward Guidance · Remi Tetot on AI, Crypto, and Uranium · IDENTIFIED FROM THE TRANSCRIPT
“So let's say you want to use ChatGPT to write economic research. If you don't have a deep understanding of economy, you cannot do it. Because it's going to spit whatever and you cannot check it. And personally, I use it to give me feedback on my work. When I write, I put my content and I ask him to give me feedback to see, you know, if I'm a strong girl or basically I'm using it as a sounding board. You know, images, that's great. But again, you know, not everyone wants to be a writer or not everyone needs to degenerate images or videos. You don't need it. It's very, very limited. Same for the music. And I've tested literally everything I could think of.”
2024-03-03 · Forward Guidance · Remi Tetot on AI, Crypto, and Uranium · IDENTIFIED FROM THE TRANSCRIPT
“I wrote about that a few months ago is how to use basically AI is not going to replace a skill you don't have. If you don't have it, you don't have it and it's going to replace it. It's not going to give you that skill to the level of a professional.”
2024-03-03 · Forward Guidance · Remi Tetot on AI, Crypto, and Uranium · IDENTIFIED FROM THE TRANSCRIPT
“Yeah, I mean, I'm with you in the point that it's not a lot of the products are not ready for primetime. And they may be used by high school students or college students to pass their classes for an exam. But if you're actually like a professional person who's, I think a professional, a human is still doing better than ChatGPT on a lot of things. Like it's, I think ChatGPT is a jack of all trades. And it's like kind of good at all things, but it's not extremely proficient in one thing yet.”
2024-03-03 · Forward Guidance · Remi Tetot on AI, Crypto, and Uranium · IDENTIFIED FROM THE TRANSCRIPT
“The sentence you're going to read every single time there is a big thing is like, oh my God, this is amazing. I can't wait to see what they're going to do in a year. People are replying that over and over and over all the time. And then they compare what was done a year ago with what is done now. The reality is people are using, I know people using chatGPT in different ways, but I would say it's a very small portion of it, people that need to use it. The product we're seeing now, I don't see it like it's not going to be used by billions of people.”
2024-03-03 · Forward Guidance · Remi Tetot on AI, Crypto, and Uranium · IDENTIFIED FROM THE TRANSCRIPT
“Use AI a lot, right? It's a big, I use, I tested also all the product you can think of to really understand. I train, you know, when COVID happened, what I did, I did an online course for a week and get trained at Stanford about AI and how to train models. I wanted to understand the technology behind it. You know, I was the tech guy at Real Vision. So for me, it was making sense to go deeper, to understand the tech, really. So here's the thing. AI is amazing. Okay, ChatGPT generating maging, amazing. Making videos amazing. Now, in terms of use case, all of that, it's actually going to be very limited. Not everyone is going to need it or use it. When you see all those Sora videos coming up last week, everybody's like.”
2024-03-03 · Forward Guidance · Remi Tetot on AI, Crypto, and Uranium · IDENTIFIED FROM THE TRANSCRIPT
“The gross in revenue is in data center, it's nowhere else. They have a bit, they have some, it's a bit flat in gaming, something else going down. The big thing is data center for them. Data center is not a recurring business that can last for decades. It's not. It's a structural thing. You know, like in 2000, a lot of investors were throwing money at broadband companies to build internet and the network and stuff. And what happened then? And all the big broadband companies end up buying the small ones.”
2024-03-03 · Forward Guidance · Remi Tetot on AI, Crypto, and Uranium · IDENTIFIED FROM THE TRANSCRIPT
“This year, I think NVIDIA have room. If the big companies have not spent the Capex money yet, they're going to reach their target on their revenues. So they're going to keep going. They're going to deliver. That's going to work. Unless we got a recession, something like a big downturn that they're going to have to review their spendings and stuff. I think this year we're good. Next year, I'm not sure. Not so sure.”
2024-03-03 · Forward Guidance · Remi Tetot on AI, Crypto, and Uranium · IDENTIFIED FROM THE TRANSCRIPT
“And they're going to be bought for cents on the dollar for the big tech by the big tech company. And, you know, eventually that's going to collapse the price of GPUs and computing power and stuff even further. This is the kind of stuff I think we're looking at in the next two, three years.”
2024-03-03 · Forward Guidance · Remi Tetot on AI, Crypto, and Uranium · IDENTIFIED FROM THE TRANSCRIPT
“I would say, like we're seeing now. And I think what's going to happen, NVIDIA is going to correct dramatically. Because if you have 75% of your revenue coming from those and at some point it just stops because everybody got their data centers, it's going to create a bit of a problem for them. And the thing I realized with Nvidia is when they talk about a fraud with CoreWave that they give them, they lend them money and they book them the GPU to open data centers and stuff. And I think I talked about it because I saw another company or two other companies doing the same with them. And I'm like, okay, what are they actually doing? It's a VC strategy. They're throwing money at company building data centers to compete with the big ones. And eventually the one that's left, they're going to have a big tech indicate. The big upside for them is that they make them buy the chips and they give them access to the chip first. So what we're going to have eventually in the next two, three years, all those small companies, like 90% are going to fail.”
2024-03-03 · Forward Guidance · Remi Tetot on AI, Crypto, and Uranium · IDENTIFIED FROM THE TRANSCRIPT
“Mach 7, or the Mach 5 in this case, you know, we keep Amazon, Apple, Microsoft, Google, and Meta. If you look at them and look at the CapEx for 2024, the Capex is going to be up 30 billion. Now, a lot of those are for data centers. Now, if you trim it down to Amazon, Microsoft and Google, which the three of them, they have data center product, cloud product, basically. They generate 120 billion a year, which is double what Nvidia is making, for example. Those companies, it's not only AI driven. They need to build bigger data centers because there is more data, you know, it's growing. But you're going to reach a point where data centered, they're going to arrive in maintenance and eventually upgrades, but not safe.”
2024-03-03 · Forward Guidance · Remi Tetot on AI, Crypto, and Uranium · IDENTIFIED FROM THE TRANSCRIPT
“It's been the darling of the stop for the last year. Now he almost became a meme stock. You know, I wrote a piece about NVIDIA when he was at 400 around the flags around it. And I did say, although there is flags, you know, when a stock trained that out and there were so many people behind it, like I would not short it. And obviously, you know, that was September last year. So I think the stock market proved me not wrong. Yeah, proved me wrong. It went flying. NVDI is making, is it forecasted to have 60 billion in revenue this year, all right? And 75% of those revenue are data center driven. Now, if you look at the...”
2024-03-03 · Forward Guidance · Remi Tetot on AI, Crypto, and Uranium · IDENTIFIED FROM THE TRANSCRIPT
“Maybe we'll go company by company in the Mag 7. Like what stands out as the most overvalued, I mean, Apple, I have no doubt that it's priced to sales multiple is significantly higher, even though it has grown. And, you know, maybe you're applying your general point to Apple in Pacific, you say, okay, the iPhone was this incredible thing. Apple Vision Pro, maybe a lot of people are going to buy it, but it's not going to be as successful as that. What do you think of as a sort of fair value for these companies based on earnings or sales?”
2024-03-03 · Forward Guidance · Remi Tetot on AI, Crypto, and Uranium · IDENTIFIED FROM THE TRANSCRIPT
“Considering the level of revenue doesn't make much sense, right? Mathematically wise. I don't see Amazon generating 6 trillion in revenue in the next 10 years, for example, unless the killed all the small shops. So when you look at it like this, it doesn't make much sense. But when you look at all the tech giants in 2000, the ratio revenue market cap was like much higher than what it is now. But again, we're talking about companies that were making 20 billion in revenue. Just Microsoft was doing 20 billion in revenue back then, or 17. And now they're doing about 140. So it's, you know, it's all in terms of absolute term now. Do we get the same kind of growth forward? I personally doubt it. It doesn't make sense to me, even if you include AI or whatever you name it. So yeah, I think it's. Interesting to look at it this way.”
2024-03-03 · Forward Guidance · Remi Tetot on AI, Crypto, and Uranium · IDENTIFIED FROM THE TRANSCRIPT
“I'm writing a piece now for next month, and I was looking at revenue, you know, the revenue market cap ratio for all the tech companies, all the big ones. And they're actually a lot more, they're more expensive than 10 years ago. Which is odd because all the revenue in the last 10 years is up 10x. Do you think that the next 10 years is going to be up to another 10x?”
2024-03-03 · Forward Guidance · Remi Tetot on AI, Crypto, and Uranium · IDENTIFIED FROM THE TRANSCRIPT
“There is room to go still on the tech side, but it's getting obviously getting very expensive now. You know, I'm still very cautious. I use the business cycle a lot. I think the business cycle is going to bounce temporarily. So we might see ISM, you know, improve, even get above 50 this month. On the longer period, I'm not sure yet. Now we're having the US is a bit weird because the stock market, now we're looking at the stock market, thinking the economy is fine when before it was those that were around, you know, the big tech, it's creating a bit of distortion. They're like 30% of the S&P or something like that, give or less. But yeah, their revenue is like 70% out of the US, for example. So people will look at the stock market thinking it's fine, the tech is fine, but the revenue is not really being generated.”
2024-03-03 · Forward Guidance · Remi Tetot on AI, Crypto, and Uranium · IDENTIFIED FROM THE TRANSCRIPT
“It's my pleasure. Remember, how are you thinking about macro in 2024? And I guess we'll start with the obvious, the stock market bull market, which is largely, to a large extent, being driven by very large cap technology stocks appreciating in price. What are your views on that and what are your observations”
2024-03-03 · Forward Guidance · Remi Tetot on AI, Crypto, and Uranium · IDENTIFIED FROM THE TRANSCRIPT