Transcription
Ron Baron, Kathy Wood, and Jensen Hong, three of the biggest names in investing, just had remarkable comments about Elon and his companies this week. Today, we go through what they said and what they did with their money.
First, legendary investor Ron Baron explains why he believes SpaceX could someday be worth 20 to 40 trillion, and why he says Baron Capital has already turned roughly a $2 billion investment into a stake worth about $25 billion. And on robots, he says nobody will ever match Tesla there. And Elon's talking about 100 million Optimus units eventually billing. Those are targets and his timeline slips, but that's a scale he's aiming at.
Second, Kathy Wood of Arc Invest bought about $14 million of SpaceX stock yesterday, $51 million of Tesla stock last week, and more of both just a few days ago.
And finally, Jensen Hong, the Nvidia CEO, called Grock, Tesla self-driving, and Optimus world-class projects. "Every single one of them is revolutionary," he says. And he says, "He wanted robots are right around the corner. Even Elon chimed in. Jensen is awesome." He says, "We're watching both the Baron and the Jensen interviews today." Let's get into it. Got Joe Bacti here with us. He's an economist, investor, and tech entrepreneur. He has a YouTube channel covering both Tesla and AGI. Check out his website and community at pionerlands.org. Welcome, Joe.
>> Great to be back.
>> We're going to watch a few videos here. Ron Baron, legendary investor, big proponent for Elon and uh Tesla and uh but he and SpaceX now of course and uh he saying some big things here. So let's listen into what he says. First he says SpaceX will be the biggest company in the world by a wide margin.
>> Buying SpaceX in 2000. So Tesla was 2014 to 16 and then they did employee tender offers uh every year twice a billion dollars and uh they get oversubscribed very quickly. And then Elon started in a private company Elon started buying some of those employee tenders for himself.
>> So great signal. So because we had a uh and he owned 44% of the company and and so so we had this good relationship with him. So we got included and we were treated the same as Saudi Arabia, as Abu Dhabi, as Fidelity. We're a tiny investor and that's how they treated us. It was incredible. And so we invested in the aggregate in 27 discrete transactions since 2017. purchased stock and and we were either the number one or number two or number three purchaser on each of his transactions in in SpaceX 27 times. So, we're talking about selling, we were purchasing. So, it's other than just these tenders, uh we told them that if they ever see any stock that is available, we want to buy it. And so, we would get these calls and say, "Here's a guy who's been in the Lester a long time. He's got $25 million for sale. Do you want it?" said, "Give me a day. I'll call you tomorrow." And they called me going back tomorrow and we want it. We bought it. So, we kept 27 transactions and uh uh and in the aggregate we invested, I think it's two billion so far. Uh and it's now worth about in June 25 billion and then we bought a billion dollars more on the IPO. And the reason we bought the billion, they said, "Well, how come that we bought a billion more because we owned 1.25% 25% of SpaceX and I didn't want to get diluted."
>> Okay, I actually played the wrong clip. We we'll get back to the clip where he said SpaceX is going to be the largest company in the world, but he went through there 27 transaction. He keeps buying. He keeps buying. In 2017, he even bought more when he had already had so much at the SpaceX IPO and of course everybody's laughing a little bit because stock fell from the IPO price 130 to 100, maybe even go down even further. What's your thinking about his strategy here?
>> I mean, he's a very long-term investor, very fundamentally oriented, and as such, he doesn't have much to lose, by the way, because he was in SpaceX so early. I don't know how much he bought at IPO, but probably not as much as he owned.
>> Not even close. So, yeah, we have to look up, but he owned a lot. So,
>> you know, it's here's the thing. It's like when you have investors like Ron Baron, you have to be a little cautious as a normal small institution investor because he's in a different game. And so going in at a billion IPO, you know, it's basically doing the IPO and Elon a favor in return for all the favors he did for him. It's not that you think it's a great price, but if you have five, I don't know how much he has a lot.
>> He had 25 billion. His two billion became 25 billion and then now he put in a billion more. Yeah. So yeah, if you have 25 billion in that stock and made a total killing, he made a killing. Going in with a billion is kind of a extremely weak signal. That's not conviction. That's the opposite. It's like not doing that would be very offensive for Elon and for the for the book runners and everyone. So basically, it's like a VC round, right? If you're in and you want to up your share or want to keep your share, your prorated, you do it because not doing so is like a bad signal. But secretly you think well, you little dummies like you invest in that round. I was in two rounds before. I'm just continuing to send the right signal. But obviously I think my investment two rounds before was much better. So you know it's it doesn't mean Ron Baron assessed the situation as oh this is a great price. I need to invest. If that would be the case he wouldn't be in with a billion but with 25 billion. So at IPO. So I think everyone who's a little sophisticated knew that IPO price is a little not a good price. Um, that was obvious especially if you're an early investor because you see the incane increase and you know strategically I agree with him that SpaceX will be the largest company very likely if they just execute on their plan because no one can do what they do and they have cornered the universe the sun that's a lot of energy so but they need to execute on it and is it tactically the right way to go in I still say no I said no at IPO I definitely I sold short calls at 210. Not a lot. Didn't make me a lot of money, but it made me 100% of the premium. And uh I still think it's too high because we have all these unlocks coming. By the way, who's getting unlocked? Ron Baron. So, no wonder that he wants the stock not to go down too much because I think he has to sell a little bit. And that's the problem with these unlocks. Even the biggest bulls might sell some shares, especially the employees because if you're an employee, you have two million bucks, you you make whatever you make a 100,000 and you have two million in shares. I mean, every wealth advisor in the world would say, well, just take some off the table.
>> Yeah.
>> So, and and that is everyone like this entire 12x or 10x of shares that coming in right now based on what is currently float is pure sales pressure. Yeah. because you can't buy. I mean, you you get unlocked. All you can do is sell. And even if it's just 10% of them being sold, it's just sales pressure on the stock. That's the problem. And so, again, that's a very tactical argument. Um, I think the thing that offsets it or on the other side of the equation is Colossus 2 and three. I in our analysis, they're coming online this year and that will probably get us a massive influx of additional contracts. Massive. I'm talking about doubling or tripling potentially their entire earnings base from 11 billion to 30 billion or something. So that's not peanuts compared to Tesla. That is the equivalent of launching 500,000 robo taxis. So that's a big chunk of money coming in from Colossus that is not really on the radar I guess. So that could drive up the stock and then they have all these analogs that drive it down. What's the net outcome? I don't know. My gut feeling it's further down even with the Colossus thing. So yeah, so basically strategically I agree with him. I just don't think it's a tactically smart entry point. I do think we will have a very good entry point this year because once it hits the bottom, it's a total buy. It might be hotter than Tesla. Who knows? Let's see what happens.
>> I remember you uh were somebody who was saying the IPO, everybody was hot on it and you said, "I'm going to wait for it because I know it's going to be very volatile," which it is. And now it's fallen significantly. So, you called that correctly. Um, yeah, I do expect that there's going to be data center uh deals, more data center deals beyond Google and Anthropic. There's going to be a big one. And I think I've been saying Microsoft and others. I think they're lining them up. They're going to be announced. We heard the All-In podcast people's, you know, a few drips in here that this is hinting that it's coming. But you're thinking that even with these kinds of deals which are huge billion-dollar deals and I think that they will lock them up not only for like terrestrial data center like now so that they can rent Colossus 2 when Colossus 2 goes online and I think it's got the B300 chips in there so it's even much more powerful.
>> um they'll lock them up also for the space data centers and that will be the story like okay
>> well but I think here's the problem with it you know in what market we are people are ruthless and crazy I told last session the micron story I mean this is all completely nuts and but even in a market that's totally nuts getting in 20 billion in earnings still counts right so it will support the stock if the stock is at 1.5 trillion you're getting in let's say 25 billion in additional earnings. You can do the math, right? That's a 40x that's a what it's a 60x price earnings. That alone would get you at a 60 price earnings to 100 1.5 trillion and then they have 10 billion existing. So, but if you want to be super critical and nasty, Micron is at 4.5 price for price earnings forward for 2026, which is nuts. So you know you could argue well if you have the same argument as for micron that you say terrestrial data centers are not the future we are like completely heavily discounting this stuff true it's 20 billion in earnings but they will not be able to continue that and the market would be right that the terrestrial buildout speed for SpaceX might accelerate one more time next year but then hit a wall with everyone else I think because the resistance is ramping up on Earth which is Like funny enough, the whole bullc case for SpaceX.
>> Well, no, but I mean it's both the bullcase for SpaceX because you already built these Colossus data centers and everyone else is getting resistance. You already built it. So, those
>> Well, there's there's rumors about Colossus 4 and that they are scouting a spot in Texas. So, if they can pull off more, that's just good. But I think these things will be heavily discounted because everyone says these are one-offs and then it stops. And to your point, I do not think that the market will price in a lot of positivity about space data centers. Okay?
>> And I do not do not think that people will sign contracts. They will sign maybe letters of intent because there is no data center in space right now. And that is exactly the problem with SpaceX that they will have this gap, the valley of tears between the promise and the reality.
>> Sure.
>> And the big question is how long is this valley? And I believe it is definitely a valley. I also believe it's shorter than people think. But I think that no one will know that until it's over. And that's my trajectory for SpaceX. I have this theory that we bought them probably around November and then we'll have a tough time for quite some time.
>> Yeah.
>> And Starship is not going to help because Starship doesn't do much in my opinion if you don't connect it to the data centers. And then people will say, well, it takes forever. it's like 2030 or something and then they will absolutely surprise to the upside probably in the second half of next year and say boom here's our AI satellite and it actually works and then they will sign so I think that's the problem like what happens in between in between it's tough
>> in between it's going to be it could be could be total gas here macro hard and digital optimist if that comes out and they somehow prove that it's equivalent to anthropics code and they can show, hey, I can actually make money out of this. And then they
>> I think you're right. Micro is the is the joker, the as
>> that could do something. But
>> I don't want to sound too bearish, but the problem with macro hard is that anthropic and open AI, that's my guess, are in dire straits when it comes to the market because now we have Kimmy out, we have Grock out, ironically. So they are under a lot of pressure where people say is that even the way to capture value. So the data centers are proven. I think the data centers is like total gold. people understand okay if you can pull that off totally game over but the model unjustifiably uh will be seen as like who knows maybe it's all goes to China and if that becomes the same becomes free right if there's so much competition
>> very smart about
>> exactly it's all open source the Chinese are going to kill anthropic and open AI and then you come with macro hard and if the market is dumb right now or then they will say well whatever it's just another anthropic so which is totally wrong by the way because you still need to make it work as an enterprise product. So I think macro hardut is super hot but will the market absolutely like you know
>> turn that into a stock story when if anthropic and openi are also are getting killed at that moment macro hard is not going to help the stock too much in my
>> understand yeah it's a good that's a good nuanced answer actually all right so Ron says that SpaceX will be the biggest company in the world by wide margin okay so that's what he believes his firm expects to make 20 30 times its money from the IPO price over the next 10 to 15 years long term. Again, it's AI data centers, right? With SpaceX reaching a 20 to 40 trillion valuation, it went public at 2 trillion. Okay, unique business. We think it's going to be 20 or 20 trillion, at least 40 trillion. I mean, he's been saying this for a long time.
Um, okay, let's talk about what he thinks about Tesla. He he believes that no one will ever match Tesla and robots. Uh, let's watch what he says here.
>> Tesla probably has a bigger advantage there than in anything they've ever done.
>> Remember what I said earlier? Their mission was never about transportation alone. It was about being in and getting us to renewable energy going into renewable resources.
>> So they can make 10 million robots a year though in out of uh I guess out of Texas. No one else can do anything near that. So like that I almost would
>> he's not talking about 10 million. He's talking about 100 million and then he's talking about a billion.
>> So is that how they year? But so nobody's going to do that. But they
>> no one's going to have any no one's going to ever do anything like his robots.
>> Yeah.
>> So So if you think about how you're able to withstand a share price going down is that the all the price does is tell you what you can buy and sell a security for in a given day. It doesn't tell you if a company's doing well or not.
>> You guys aren't getting signal from the fact that the stock is up or down on
>> don't give signals. They just know there are guys who trade and systemic investing. They trade and they make stocks go up and down. They follow the leader down, they follow the leader up. You know, that's not what we do. We try to find businesses in which we can invest.
>> Thoughts of what he said?
>> Yeah, I mean, he's totally right. That's how you make money. Um, and yeah, I think it's like the uh the idea to find a business then different from Warren Buffett, predict the future, which is not that hard in my opinion, not the stock price future, but the business future. And that's what Ron Baron does and say look this thing is undervalued and then buy it and hold it and make money is a very good model and it works with SpaceX and with Tesla.
>> I think the the key thing is like u I I like what you said about if it's software macro harp and then Chinese strategy [clears throat] is that because they're second their second player right now they they don't want anthropic and open eye to run away with it. So they're going to go, I'm going to give everything away as open source. If it's open source, then it then it challenges Anthropic OpenAI and even SpaceX AI's ability to charge money for these software because they get them as almost as good and then
>> well the interesting thing is like the interesting thing [clears throat]
>> let me finish what I was going to say though but that's that's that is a good nuance concept. But if it's physical AI where you have to make millions of cars to make them transportation or you have to make millions of robots, now you're playing in a different game. Now you're playing with others who can actually make that many robots. And what these guys are saying is there's going to be few maybe there will be just like just like EVs. There'll be Tesla number one and there'll be nine other Chinese companies behind it as competitors. And that might be what happens. You know, uh the the competitors will be lower in ability to catch up, but they will always be the virus because no one right now is building 10 a 10 million per line uh factory. Uh and they're already every month that goes by, they're already a month or two ahead of all the others.
>> Yeah. But I think both for micro hard and for the robots and the robot axis, I think that vertical integration is extremely important. And when you look at vertically integrated players, there is none other than Tesla and SpaceX.
>> Like who is a frontier model producer that also has an enterprise product with cursor that also has data centers and
>> that also has the ability to manufacture robots. Yeah.
>> And yeah, that's one stack. And then the robot stack. Exactly. Who has the model for the robot? Who has the data centers to train these mod? and who has the robots and can manage no one and so no one at scale and so that will turn out to be a very powerful mode even against the Chinese right because the Chinese also don't have that the Chinese have overwhelming manufacturing power so if it if it's just Tesla versus the Chinese on manufacturing I would actually bet on the Chinese because you cannot beat the Chinese but it's actually about vertical
>> no they need to copy Tesla though that's what
>> yeah but then you know how it goes so Um but in terms of vertical integration, I don't think anyone can do it in China right now.
>> That's a very good point
>> and because this vertical integration even beats open uh open weights and everything because you need to optimize even the digital optimist needs to be optimized. It's not just robots also even macro hard. So I'm not to your point I'm not against macro hard as a super valuable thing. I'm just saying tactically it could get into difficult waters with enthropic and openi getting killed potentially or getting into trouble with the market but strategically I think in a way macro hard plays on the same strength as cyber cap and optimus because it's also a vertically integrated stack you need to optimize this entire thing and we see it now with cursor and these massive leaps elon just confirmed August 7th that's very soon a release of gro 4.6 six and then 4.7 coming very soon after. So that's a massive rollout speed for the new Gro models.
>> Yeah, we got a couple more things to cover, but I just want to you just came up with a thought that I've been thinking about recently. The new angle, there's two things that I think Elon has up his sleeve of what will make him different than the current race with LLMs. Well, right now in order to get a brilliant AI, you have to train it as much data as possible, as much data center as possible. You train it. Now you got this intelligent model. Ilia Suskov, who left OpenAI, has formed a company called SSI and he just got a substantial agreement with Nvidia, right? And what his the [clears throat] theory is of what he's built is a learning AI one that is like I think he called it to a 15-year-old. It doesn't have to be like this thing was trained with everything. No, what what he's optimized for is this thing that doesn't know what's going on but is able to learn and learn and then learn and just it's like optimized for learning. And if you do that, this thing will surpass everyone else. That's the theory. And uh Elon hinted uh saying something like that is where they're headed. He goes I think where we're going there's no need for training data. And I think that that applies to both the bot and the LLM where they're not you don't Yes, you want to train it, but the whole point is optimize it for learning and be able to then figure out how to just on its own figure out things. So that's one. And the other one I'm still um trying to understand and talking to experts on is is there truth to Elon's and Ashok's comments at the earnings call that vision only the same technology that they use for full self-driving can create a full self computer. what he call it full self computer right which is you know the you look at the screen and you will learn to understand what's going on and then be able to act immediately within seconds as opposed to uh right now I can use anthropic they just launched a feature where they go show me a video of you using your screen whatever if it's the website whether it's Excel spreadsheet whether it's a custom app that I'm using show me and tell me verbally teach me what you want me to do click here click there this is where I've grab that data then it'll learn it and it'll do it. What's the difference between the two? one is much more immediate. But I think if that full self computer vision in and then immediate action out because it's got the context and it's got ability to learn on its own, that might be the difference. But I don't
>> Well, I think the bottom line is that there might be ex enormous algorithmic improvements and paradigm shifts coming up soon because I mean obviously it's worth coming up with that. By the way, the human equivalent uh for people who are interested in the brain, the actual training doesn't happen with the human brain. The training of the human brain happened genetically, right? Evolutionary.
>> Yeah.
>> So our weights are given with us as babies. So the babies,
>> right? They just learn based on the weights.
>> So the weights basically mean you can look around and and the baby sees, oh, this is like a animal or something. Even though they never learned that, they see it's a cat. So all the weights oh the background movement this is all genetically programmed in. So that's an the training of the human brain happen through evolution. So everything else is inference and that's not how that's also why a baby can immediately understand the difference not with a million cats but one cats like oh cat dog different doesn't need this stuff. So that I knows that and of course in memory right now we see deterioration with more context right the mods get worse and worse with the chat length. It's like oh now I for what they get tend to get worse they should get better and better and so they if you fix that that has enormous implications but I think whatever happens these they will then be copied I mean these new algorithms will also be copied by China and so on so it's not that you have a monopoly
>> that's that's the that's what Tesla is saying that just like you're just like you somebody says oh you know full robo taxis can be copied Well, in order for Tesla to get there, they had to have supercomputers. They had to have all that data. They had to have that training. They had to do the full self-driving where you got vision in to action out. It's not Yes, you can copy it, but you'll take you you'll be years behind to get all the edge cases to get all that always be had. That's their argument.
>> Well, it depends on the nature of that algorithm, right? It depends what how it exactly works. But
>> but LLM wise, they just copy it, right? And that's that seems to be correct. They are copying. Exactly. So, so uh the bottom line is this will be a crazy race and that's why I'm not bullish. I'm very skeptical actually with anthropic and open AI. If you have a single layer thing, they can get killed overnight. If something like this happens,
>> they could be dead. Whereas vertically integrated stacks, the more you are hardware or infrastructure, the more you have a giant moat. And the more you do both, the more you win because then you can switch out this thing, deliver the full stack value like macro hard but remain flexible on the top and not you know built into your own thing because you have the full stack you know okay I can deliver this whole thing. So you also see that gro is a nonopen model that prices itself and behaves a little bit like an open model. It's much closer to Kim K3 in the cost side.
>> Well because they they give it to low cost. Yeah, it's brilliant. Okay, so let's a few more things here. Ron Baron actually called Elon the best builder on the planet. Okay, and that is important. I we just talked about it
>> holding himself out as I am a builder. I am the best builder on the planet probably. So when other people take two and three years to build a data center and cost them 40 or 50 billion dollars, uh then we build it for a gig. uh then we build it uh in six months 122 days six months we build it and it cost us a fraction of what it cost them
>> and not just data centers um and not just speed right so there is getting started first like optimist factories but also how fast they can build that people forget that that's both the case so once maybe the Chinese can copy like you said in their speed of building these things but no one else can that's one of the things
>> yeah never underestimate like the power of execution and um I'm a little bit more I think I wouldn't call Elon the best worst best builder because I think I always have great respect for the Chinese. I mean, they're killing everyone. They'll outbuild everyone. Even Elon says that Elon is more like he is sufficiently good at building and in the west might be the best but he combines it with something that the Chinese don't have and that is like visionary first principle. That's where the Chinese are very bad by the way. He cuts out steps and so then
>> yeah no matter how fast
>> to do something that the Chinese absolutely are bad at maybe they become good at it but right now they absolutely cannot do that and that's like imagining something completely new that doesn't exist based on first principle on theory that's the weak the Chinese strength is pragmatism and super engineering and then being very good at math but they have a problem in like saying okay what if something that just doesn't exist we make it up with theory and say this is what it is and then we pull the trigger and execute on thing that's a very western concept and then Elon has mastered that and then Elon different from other people who master that
>> can also execute because you have other people who are great theorists and even are right uh but then you know they don't
>> followers come in and they follow fast and then they wipe you out but Elon is always on to the next thing and uh he's he's just a grill so Kathy Wood uh Arinvest she bought $14 million of SpaceX yesterday and then she's keeps buying the stock. This is just her beginning. She bought uh on Monday uh 4.4 million shares combined. Second big buy in a week, 15 million added just days earlier. As we talked about, stocks falling nearly 50% from his 225 high. 13th decline in the last 16 days. She calls a sell off a wall of worry arguing bull markets don't end this way. They end when everyone believes the sky's is the limit. So, it's going to go back up again. two catalyst loom SpaceX first earnings report. Uh but of course one thing he didn't mention here is that the lockup is coming up too. That's a negative. So this is the uh she's also been buying Tesla. Thank you to Alexander Mertz uh both Tesla and SpaceX dollar cost averaging as uh the stock falls that she'll keep buying it. So that's a that's a good sign that she's doing that. Thoughts on Arc Invest?
>> Yeah, I think also big picture on SpaceX is I'm just bearish because of this unlocks and but from a fundamental perspective, we are now in a very interesting buying territory. I think if you are interested in a position, now is the time to initiate the position and then dollar cost average and I think can we hit in a in a panic or something 50 or something absolutely but that would not last very long. I think now we are really scratching the bottom. Yeah, the bottom can fall out for like a month or a week and then we can drop to wherever. But I think there is now massive resistance coming on the buy side even with unlocks because we have colossus coming and the market won't be that negative forever. So I think it's probably a smart time if you're a little more stretched out to start looking into that. I'm still waiting because I'm pretty sure it gets lower at some point. But we are getting into very, you know, you have to really pay, if you're not getting in now, you have to really pay attention and catch it because otherwise you might be late. So I think it's getting very uh interesting.
>> And then we got Jensen Hong here. Let's listen to what he just said recently in an interview.
>> Tesla and XAI. Uh Elon's Elon's, as you know, is just an extraordinary engineer. Um and I love working with him. Uh we've built some amazing computers together. We're going to build many more computers together. and and uh uh the work that he's doing in Grock uh his self-driving car uh um his uh Optimus uh these are these are all every single one of them world class every single one of them revolutionary every single one of them are going to be gigantic opportunities and and um we're delighted I'm delighted uh to be working with him on that and and so I I think the um uh the the optimist opportunity is just right around the corner it's very likely that human robots are are going to be robots that we can deploy into uh the world relatively easily and uh uh this is the first robot that really has a chance to achieve the high volume and technology scale necessary to advance technology and and so I I think this is likely to be the next multi- trillion dollar industry. I'm very excited for him.
>> And Elon said Jensen is awesome. So, of course, uh, Jensen Wong, he partners with everybody. He loves everybody because that he sells the a product that everybody needs and so he's supporting everybody. But still, it's good to hear.
>> But it's normally I just say he's a cheerleader. He's the president of AI. So, he's like loves everyone. But he was quite specific here and that that's interesting that he was specific saying Optimus is the one single platform that's most likely to push this all forward. So, that was a little more than normal. And I think Optimus is totally underrated. People are, you know, they're like, "Oh, Optimus is dead. We don't we didn't hear anything." It's very very clear that they have solved the Optimus problem in the first commercial iteration. In my opinion, it's it's totally clear. They're building out a massive factory. They're not saying a word. Elon said exactly why because it's the final product that actually launches commercially and they don't want anyone to see it before launch. So I think um but again um I'm actually very bullish on Tesla. I think Optimus is also a big now getting into very relevant territory timing wise. Probably this year or something we will see some serious stuff. It's not going to be a demo.
>> I think they're going to launch this thing. I don't know if this year or not, but they're not going to do a demo. They are going to say, "Okay, boom. Here is Optimus." uh when they present it and um the problem that I have is the market sentiment even this this wouldn't help them right now because the market's like dude we are the 10 we need to see at least 50 billion in earnings before we even
>> even you know move the stock 10 bucks. So that's that's a hard insurmountable problem right now. So that sentiment has to shift before I'm getting more bullish on the stock because if that's the case okay fine what are you going to do? you have to wait until the tantrum is over. But if the tantrum is over, I mean, I think, you know, when I look at Tesla and how everyone behaves, I think these people are totally killing it. They're like on track of launching all these things. I think people are a little bit like delusional right now. Like clearly OP like hello, they're building the entire Optimus F mass production of Optimus putting into the ground right now.
>> Yeah. So again, so I don't I want people to I did a whole deep dive on this and I just want people to hold be careful because don't expect that Optimus is going to make a dent this year, maybe not next year because it's going to follow the same pattern, the same track uh playbook that Robo Taxi did. What they will do is hold everything, don't show you anything, no progress for years until it's ready to all go. What they're going to do is they're going to build a thousand robots. They're going to have a thousand, maybe 5,000 this year, but they're all going to go into the university to be trained because they could, if they wanted to today, roll out an Optimus and sell it to customers for pick and place. They can do that. That's not their interest. What they want is just like you have full self-driving that can drive in any part of the world. They want a generalized robot that can do any task at any time just by training it. To get there, you need to take your 10,000 20,000 robots that they said, put into university, train it so that it will learn and it'll look at something, you show it once, it'll get it. Any task anywhere. Then, uh, when they roll this out, it'll all go to their own factories and maybe year later, two years later, when they're ready to go with the Optimus version 4, that's the one that I expect that they're going to sell to customers. That's the one that's going to go big. So, it's gonna be nothing.
>> Nothing. And then all of a sudden, bang.
>> But it has to be generalized.
>> They're aiming at a new Optimus every year or something.
>> Yeah. So, you'll have Optimus 3, but that Optimus version three is for training. It's for university training.
>> Yeah, but that means they have Optimus version four next year.
>> Yes. Option version four is next year. I agree. Mid next year. But when that comes in the mid next year, that's the one that's going to be
>> Yeah. I mean, that's not that much. Yeah. Right. That's roughly what I'm saying too because here's the difference with cybercap. I think what people don't understand about cybercap is there is no competition in cyber cap. There is just no one who can do that. Maybe China in 2028 or 9. With the robots it's very different. There is like massive competition and these people are extremely serious from Hyundai all the way to uh to figure one of course and to electronic and and to everyone and and China. And so they Tesla has to
>> has to roll out something next year.
>> You're right. Good point. Great nuance. Again, that's a difference. You're right. That's right.
>> That is the other reason why you're absolutely right. Why they've slowed Robo Taxi is because well, if the Whimo is not really a competitor and they want to do it right and if they rolled it out too soon, then they might get, you know, bad press all that. So, let's wait until we nail this down. It's like what is the there is no there is absolutely no upside accelerating this beyond reasonable and there's massive downside on optimist it's the reverse right if they miss out and the Chinese are in all the companies and have pilots running all over the place they're building relationships and everyone on the Chinese roads
>> you are having a real problem so Tesla needs to nail this and really do a blitz attack once they're ready and and not too early but definitely Definitely not too late. So this is a much tighter timing. That's why I think next year is obvious.
>> Interesting. Okay. Well, that's why I like talking to you because I think you've just
>> And by the way, Chris, dumb money. He he he knows that stuff very up close. He just went to Upronic. He reported back and said, "This is amazing. I saw the latest robot. Mind blown."
>> And so that means Tesla also mind blown. No one knows that. But 2027 is going to be the robot year.
>> It will be. Thank you. Thank you so much, Joe. Follow him at pinerlands.org. Thank you.
>> Thank you.