Transcription
I sent you a year or two ago an article about a young man. It was an interview in Baron's, and he was 33 at the time, and he'd become a portfolio manager. He lost his legs to meningeal bacteria. And I said, "Is there anything we can do to get him out of the wheelchair?" And you said, "Yes, there is. In three or four years, we can give him an Optimus body, and then we can use transistors in his head, in his brain, to let him function as a normal person, dance and sing and walk and run." Have we been able to make progress in that area?
>> Yes. So that's a compliment of two of my companies. One is being Neuralink, and the other being Tesla. So Neuralink is also making good progress. Who now has, I think, over 10 patients with Neuralink implants. And these people who didn't have the ability to move their arms or legs, in some cases were completely locked in, like like Steve Ky. And they can now communicate, I think, as quickly, almost as quickly as we're communicating right now, which is very cool. And that's, that's going to continue to accelerate. What we can do is use a Neuralink implant that is taking signals from the motor cortex brain and also receiving signals from the somatosensory cortex. You have someone who's lost their legs, Optimus legs. And so you, I mean, we're really getting like the $6 million man here. I mean, from back in the day. I don't know if you watched that show, but I watched it.
>> I, I watched it.
>> Yeah, I thought it was pretty fun. We can actually give someone superhuman cyborg capabilities like the $6 million man, but less than $6 million in this day and age. I mean, $6 million back then was a fortune. These days it's like nothing, but much less than that. I mean, like for something that would be reasonable and affordable, it might be well, like $60,000 type of thing. And you could take the signals from Neuralink, the mind, that would be transmitting to the legs, and transmit those to the attached Optimus robot legs. You would actually be able to run faster than a human.
>> Just like $6 million.
>> You, let's switch to XAI. So, three years ago, you were here, and you had either just purchased or about to purchase Twitter, which you've renamed X, and you were widely criticized for that. And
>> Yes, I was.
>> Right. In fact, you, you even mentioned here on stage that you didn't want to have anyone else be angry at you because you had enough people trying to kill you already.
>> Yeah, totally.
>> So, you were buying X, and it was $42 billion, I think, and you were in the process of raising money. And I called you up and I said, "I would like to invest with you $100 million in this." It was $60 million for one of our funds and $40 million for me. And you said, "Really?" Said, "Yeah." He said, "Really?" And I said, "Yeah." And you told me that you thought I would make a double. And I said, "Well, I hope so." But to me, it felt like you made us $8 billion in Tesla, and it'd be not very appropriate if I didn't support this new venture that you were doing. So, we invested. And then the day that we paid the money, we marked it down 70%. So, man, this is how I know you're a true friend, Ron, because I, I do regard you a true and trusted friend.
>> I got more to the story.
>> The test of friendship is who supports you when the chips are down and the times are tough and everyone's against you. That's a real friend, and that's you, Ron.
>> Thank you. So, we did invest about 100 and marked down to 30. And then about a year later, we started getting phone calls from hedge funds who always seem to know things we're not supposed to know.
>> How do they do that?
>> Right? And they started saying, "I'd like to buy your stock for what you paid for it." And I said, "I think I'd rather wait and stay in," which I did. And then change the name to X and change the configuration of business. You bought Twitter, and it came together, and a social network along with this, and all of a sudden we have a business that has incredible data. Did you buy this for the data that no one knew about? The data with 600 million people talking with each other? So physical data that no one else has. And then you started Grok, which is based on our data, and everyone else doesn't have that. They got digital stuff. And then when they have Grok, then we need more data centers, and you're building those. And in a space of a very short time, less than months, you built a data center which is four times what is large as anyone else on the planet, 25,000 what other people had in in CPUs, and we built it in GPUs, 100,000 more powerful. And then, you know, now we have now we're going to go to hundreds of thousands. But the bottom line is the investment that we made, we put up more money, and we have a total of $350 million invested over the past 2 or 3 years, and now it's worth $700 million. So everything you touch is like that. It's the most unbelievable thing I've ever seen. So everyone is invested in technology, and we are investing in the technology person, the best engineer on the planet. So thank you very much. So the vision is, did you buy X? Did you buy Twitter because of the data? Is that, did you have all this in your head before you did it?
>> Not really. I, I just bought Twitter because I thought it was having a negative effect on civilization. Just sort of pushing ideas that were anti-civilizational. You know, it's sort of having some captured by the far left. That's, you know, it's fair to say the radical that they would want, wouldn't regard themselves as such, but it was captured by a group of people whose political beliefs are those of deeply San Francisco or Berkeley, which is about as radical as you get in America. And it wasn't a good forum for debate because they suspended many people on the right, including the president, the sitting president, which is really unprecedented. I think we need to have a public square where there's true freedom of speech. Freedom of speech is the bedrock of democracy. If there's not freedom of speech, people cannot make an informed vote. And if you cannot make an informed vote, you don't have a real democracy. So the purpose of acquiring Twitter was to try to bring it more to the center. No left-wing voices have been banned or suppressed. What we're trying to do is give equal weight to all parts of the country so they can be a public town square where people can exchange ideas and hopefully not present violence. I think it's one of the, like said, rock of democracy. It's why it's the First Amendment, 'cause people came from countries where if they could be killed or imprisoned what they said. And in fact, this has happened all around the world. It, it's the civilizational risk had to be addressed. If America is not strong, then what to businesses matter? The America is a central pillar that holds up Western civilization. And if that pillar falls, everything falls.
So you were one of the founders, the two founders of ChatGPT and you know, OpenAI. And you had a disagreement. And it was founded as a charity. And it was your idea that you wanted to make sure that, you know, freedom of speech and all the things that you deem important for good lives on our planet were followed. Safety. And the other founder, what he tried to do and did was accomplished is that he got control. Even though it was your money, and he got control. And and you, and he said, "Elon, I'd like you to stay." And you said, "I want to go. I don't want to be a part of this." And and he offered you some ownership. And you said, "I don't want it." I said, "Here, you walk away from an ownership of ChatGPT." So you're obviously not doing all this stuff for money. I mean, you are, but I mean, it's that's, but I mean, if you, if it were only about money, you would have never left with something that's worth $500 billion dollars by itself. And so here you're forming this new entity to accomplish what you wanted ChatGPT to accomplish. But you think that we have an advantage in this because of the data, because of the compute, because of what? What are you going to do with this ultimately? You talk about connecting physical world to digital. What does that mean?
>> Yeah. Well, just going back to OpenAI for a second. The reason I founded OpenAI was because I was concerned, based on my conversations with Larry Page, who used to be a close friend of mine, that he was not sufficiently concerned about the dangers of AI. This really came to a head when at my birthday party, he, in front of a large group of people, called me specious, or favoring humanity over computers. That's troubling. Like, Mary, what side are you on? Sounds like you're on the side of the computers, but you really need to be on team humanity here, you know. So after that, I was like, "Okay, this is it. We've got to have some counterbalance to Google because Larry doesn't seem to care if humans are not." So I thought, "What's the opposite of Google?" It would be an open-source nonprofit. And that's where the word "open" in OpenAI comes from. It means open source. And and I, I provided, I provided all the money, like one of the Series A, B, C rounds, and recruited the key people like Ilya Sutskever, and bought them everything I know. And, you know, I actually even got them the deal with Microsoft, with with Satya Nadella. I got so to donate some time from Azure. And for all that, I did not seek any financial reward whatsoever. Or the reason I actually turned down the offer for shares is because I felt like nonprofits are supposed to have shares. At least last time I checked, nonprofits are not supposed to self-enrichment. That's why I turned down the offer of shares because it didn't seem morally or legally sensible. With XAI, we are starting late. With XAI, you know, we're only two and a half years old, basically. They're starting from behind. We also would have been pretty good with technology. So I don't want to pat myself on the back here, but I'm pretty good with technology, and we are advancing faster than anyone. So I think for technology ventures, the winner ultimately is the one that is able to move past this. We think that we're optimizing for the best technology, and we're doing something different than others in a digital world, and we're physical to digital with movements and visual, and other people can't match that. And also, we have the real-time data. What does that mean? Why should we do better than everyone else? Why are we going to win? But why are we going to at least be different than everyone else? So we have a really strong business. Well, first of all, I think in terms of being a strong business, I actually not too worried about even a small player that is not will be worth a lot because they will contribute so much in productivity to the economy. So it's actually pretty easy to achieve. Not pretty easy, but I mean, there will be many companies that are worth sustainably several hundred billion sustainably. So then the question of like, well, how do we achieve the lead? That comes down to three things. Are you able to attract the best talent? Are you able to bring the most amount of AI hardware online? Can you bring GPUs online faster than anyone else? We've already demonstrated that we can do that. Jensen Huang himself said that he was blown away by how fast XAI launched the data center.
>> Jensen said there's only one human on the planet who could have done that.
>> Yes, he does.
>> Speak about.
>> I keep telling people I'm an alien, but he believes me. I mean, when I got my green card, it said alien registration card. So, I mean, you know, I have the government. So let's say I've got some, some like relatively skills these days in America of of getting hardware built. If you look at the biggest successes in manufacturing in America since World War II, by far, are Tesla and SpaceX. So the idea of connecting physical and digital is that that's different than digital digital figuring out who wants to buy what. We're doing something entirely different. Is that fair?
>> I wouldn't say we're doing something. We're doing something same things that are different. If you're just saying the elements that define success for an AI company are going to be on the talent, do the hardware, how much hardware can you bring to bear? That's actually a very big deal, and we've shown that we're the best at doing that at XAI. And then third, unique access to data. And for that, we've got the X system, formerly Motor System, which is by far the best source of real-time data in the world. Those are some pretty significant assets. Come up with some very innovative ideas, but I have more ideas in my head than I know what to do with, frankly. I think we'll make some moves that are not on the chessboard that people do. So, creative moves. I should point out Grok right now, heavily is still the smartest AI, best of my knowledge. Grok for heavy is where we spawn several agents. They work in parallel. They compare their output like a study group and give you the final conclusion. And it keeps getting better. And now we've done training on Grok 5. Grok 5, I think, will be the smartest AI in the world by a significant margin on on every metric without exception. I might be wrong, but I think that will be the case. And that will be in some Grok 5. Yes, 5 is the first time where I thought, well, we have a non-zero chance of achieving artificial general intelligence. Um, not that it's a high chance. I calculate like 10%. But that's what my biological neurology comes up with, which still means 90% chance that we don't. To be fair, but I've never thought that before. And so for the first time, I think like, wow, this really could be general intelligence. At least a small chance, we'll really be something special. And and it'll be both extremely and extremely fast. So, one of the things that we're doing that I think is interesting is Grokpedia, which we're going to rename down the road to to be Encyclopedia Galactica of Isaac Asimov and Douglas Adams. The idea of Galactica is to create an open-source repository of all knowledge, like a distillation of all knowledge. And and open source meaning anyone can access it, anyone can use it. And if other people want to train on it, they can do so. Create copies of this and distribute these copies throughout, and even put them on the moon and Mars and out in deep space. And in a way, sort of a modern-day Library of Alexandria. It was the great tragedy that the Library of Alexandria burned or was burnt down. So in order to preserve this knowledge, I think we want to literally etch it in stone and distribute it widely. So in worst-case scenario, you're just going to see what we learned and pick things up from there.
>> So is there a major breakthrough that you can describe that allows us to do this with Grok 5? Is it just speed? Is it more compute and therefore we have more, you know, information we can train on? What is the breakthrough that allows us to to have this 10% chance for AGI? Is there a breakthrough or is just speed and access to data?
>> There's a couple things. It will be the largest model, best by knowledge. So this is a six trillion parameter model, whereas Grok 3 and 4 are based on a 3 billion parameter model. More this trillion parameters will have a much higher intelligence density per gigabyte than Grok. Like this is an important metric to think about, intelligence per gigabyte and intelligence per trillion operations. Quality of the data that we're training on with Grok 5 is mission critical. It's also multimodal. So it's text, pictures, video, audio. It's got to be much better at tool use, and in fact, reusing tools to be more effective at answering questions and understanding its vision will be extremely good. Don't have real-time, which is I think a really fundamentally important thing. Only other AIs can understand real-time. And I think if we can't do that, which humans can obviously do, you you really can't achieve AGI. There's some special sauce items that I can't talk about in public, obviously. You can't give away all the secrets here, just between us. But we have a few other special things that are in the works for Grok 5. So it's, I feel sentient. So when we're Grok 5, and you're talking about the advances, there is no limit. So when we're Grok 5 is better than Grok 4, which is better than Grok 3, which keeps going. Once we get to sentient levels, we go two sentient, five sentient, 10 sentient, a million.
>> Correct. The sentience will grow. I mean, what's really mind-blowing is, can how far can the sentience grow? To your point, how far does it go? I think it goes immensely, almost incomprehensibly far. The, like, like we see a path to putting 100 gigawatts per year of solar-powered AI satellites into orbit and having this be actually the lowest cost way to power and operate AI at a very large scale. For reference, the United States consumes roughly 460 gigawatts on average per year, because the average power load in the US is 460.
>> The whole country.
>> All electricity of most in the US.
>> And you're talking about 100 being in.
>> Well, roughly a electricity output. We have a plan mapped out to do that. It gets crazy. So there's what, a trillion planets like Earth, the solar system, whatever you call it, a trillion. And in that trillion, so the Big Bang was 14 billion years ago, 13 and a half billion years ago. And planet, our planet's only 4 billion years old. So there must be other planets that are like ours with all the minerals, oxygen, hydrogen, silicon, carbon. So life here has been extinguished four times. And presumably, you feel that other places, other civilizations, other planets, then we'll get off of this existence. And are they planets where the beings there are part human or part carbon and part metal?
>> Well, I think we'd like to find out. I'd like to find out. I mean, my philosophy is one of curiosity. I, I just want to know what's what's going on in this universe. Is the standard physics correct about the beginning of the universe? Are there other alien civilizations? Can we talk to them? And what questions should we be asking about reality that we don't know to ask? So that's my goal is to expand consciousness to better understand the universe.
>> So let's go to away from the universe back to Tesla again. And so you said that
>> Got to be backing down here.
>> Back on ground. And so you said that our expertise is in making things better, faster, cheaper than other people. And when I started investing in Tesla, when we started investing in Tesla, you were telling us that that it's the machine that makes the machine that's most important. The machine that makes. So you went into machine learning, machine technology then, which is 15 years ago. And and now the average car, I think is 50 seconds, 40 seconds, 60 seconds, and we're now 35 seconds. Every 35 seconds a car rolls off. And then you say that we're going to get down to 10 seconds. And you said it's a possibility we can go to five.
>> Every car is rolling off a line. How's that happen? I, I certainly see a path to achieving a roughly 5,000 milliseconds or 5 seconds, which is only really walking speed. That's like sort of a fast walk. 1 meter per second is a fast walk. The car is less than 5 meters long. So 5-second cycle time, the cars will be exiting the line at walking speed. You can run away from them. It's not going to be like they're coming out like bullets or something. As a rough rule of thumb, it's 10,000 minutes in a week. If you run a 24/7 operation and you get, let's say, 10 cars per minute, and you get, you know, let's say 10 cars per minute, you got 100,000 cars a week. So, but the question is, how come we're able to do this? The other people just not care? Or they think that, gee, if I do something, if I have this idea and we try to implement it, and it doesn't work, then I'm not going to get promoted, or I can get fired, or I'll get blamed. And if it works, then I'm going to have to do a lot of extra work that I wouldn't know how to do if it didn't work. So why don't other people have, you know, a mindset of making things better? What the, the Chinese, they've been great at copying us, and in some instances, probably even done better than us after they've copied for the first time. But how, how come do you think that other people haven't been able to make the advances we have? And even the guy from Ford just recently said, "The people in, and the Chinese have copied us, and those people in China are doing great, which is really compliments us, because the reason they're doing great is because of us." Why doesn't he do that?
>> Most companies make incremental, you know, the management team wants to do, I don't know, 5% maybe 10% better than last year, as opposed to take big risks that would fail. Obviously, I don't have a problem with risks. Yeah, I say, and I use physics to analyze things. You know, when I was in the factory one night, and I was looking at the factory, and I was like, this could be much more efficient, be much faster. The, I was trying to just rough math, trying to calculate the volumetric efficiency of the factory, to divide the factory into cubic meters and say, how many cubic meters are doing something useful? And it's a surprisingly small percentage. The volumetric density is not very good. And then the speed of the cars and the parts moving is quite slow, generally limited by the speed at which people can say, attach brake lights or a seat or something like that. So you densify the factory and improve the volumetric efficiency, which is helpful for production efficiency, because things have less distance to move. Just like a chip. You densify a circuit that's in a chip, get more efficient. You have to think of a factory like a chip. How do you make a chip faster? Well, you, you bring circuits closer together, make them smaller, and increase the clock speed.
>> So, Robin told me that you told her once that, "I think of myself as a bit. And if I'm a bit, how would I like to travel?"
>> That makes sense. A bit, or an or I say, "What's the journey of the bit? What am I doing?" If this journey doesn't make sense, I need to fix it. If the journey of the ATM in the factory doesn't make sense, I need to fix it.
>> So, when you say you're working on weekends, I spend all my Sundays working on a ship. What does that mean? What do you do?
>> Saturdays, but some sometimes Sundays, actually, just recently has been Sundays too. Yeah. Just the, the, the AI5 chip, which is going to be a great chip for Tesla engines. On that chip, that's that's the chip that goes, that will go into our next generation of self-driving cars, and it's also essential for the Optimus robot. So that chip program was in bad shape. It, it wasn't, it wasn't closing because it's, it's it's quite an ambitious chip design, and and it really wasn't on a path to success. And they also had the Dojo program, which was, it was also doing okay, but not on a path to be competitive. So I collapsed the two programs into just one program, is get everyone focused on chip, which is essential. Continue to use video for training, but we need the AI internship, which is it's a very powerful chip, but it's also a very low power. It doesn't use a lot of power. So this performance per watt is extremely good. We think, you know, probably going to be at least two to three times better than Nvidia in performance per watt at the inference level in the car and the robot. 10% of the cost of an Nvidia chips. So these are very important numbers to achieve. And I had to get the chip program back on track. So invest amount of time into it. And at this point, I have the entire physical design of the chip laid out in memory. I can visualize the whole thing.
>> But when you're talking about
>> Chip manufacturing, you say we might have to build, presumably we would have a partner with TSMC or with Samsung. We wouldn't do this by ourselves. And those guys have been doing this for. We would do it by ourselves. And and these are things that cost 20 million, 30, 40 billion. And where the people come from to do this? How could we not have partners? Well, it's only 10 years. There's not going to be enough chips in the world to accomplish what we're trying to do.
>> Yeah. First of all, I have much respect for TSMC and Samsung. And we, we've worked with both TSMC and Samsung at Tesla and at SpaceX. So TSMC and Samsung are great companies, and we want them to make our chips as quickly as they can and scale up to as high as possible volume that they're comfortable doing. It, it doesn't appear to be fast enough. When I ask, "How long will it take to from start to finish to get a new chip fab built?" They tell me five years production. I'm like, "Five years to me isn't my timeline's."
>> To me too, by the way.
>> One year to a year at a year three scale, it reels to infinity.
>> So So we can't even see three years. So then I'm like, damn. Okay, this is this is not going to be fast enough. So now, if they change their minds and say, "Yeah, they want to provide us with, you know, 100, 20 billion chips a year in the time frame that we need," that's great. But how could they not when they know that we're demanding and we are going to use the product and our own product? How could they not want to be our supplier? Or how could they not want to be partners with us? I don't understand.
>> What our partners were using both TS.
>> I mean, to really expand capacity tremendously, why don't they do that?
>> From their standpoint, they are, because we'll be using TSMC, Taiwan, TSMC, Taiwan, TSMC, Arizona, Korea, Samsung, and the Texas, back Samsung, four fabs. And, you know, from their standpoint, they're moving like I'm just saying that nonetheless, it would be a limiting factor of old rice. They're going as fast as they can. From this standpoint, there's a penalty. They're just never someone with a company emergency. That's it. Might just be that the only way to get to scale at the rate that we want to get scale is to build up a real big and output of optimized self-driving cars.
>> Choice.
>> Obviously, you're not going to drive.
>> Those are two choices.
>> To go to FSD. We don't have too much more time. But FSD, full self-driving. The, you know, we're making these tremendous breakthroughs, it seems. And you said recently that you didn't want to expand capacity for making cars until you were convinced that that was the case. You are now convinced that's the case. And again, here we're making these exponential leaps in these cars. And you said that a very large percentage of people who actually paid for full self-driving don't use it, have never tried it before.
>> Yes, it's pretty well.
>> Yeah. How?
>> Yeah. How we doing? How?
>> Something you want to try it?
>> It's amazing. And so we're now kind of insisting with customers for safety reasons that we demonstrate full self-driving because the numbers are unequivocal at scale that with with now over 10 billion miles driven, that it's four times safer on full self-driving than than not. So it's actually a big improvement in safety. At this point, we're just insisting that we at least demonstrate self-driving to customers so they know how to use it and turn it on for safety reasons.
>> So, full self-driving drives just like a person as opposed to having to code to look for every circumstance that would happen or someone walking his dog while he's driving along. You have to identify everything with a code. What AI does, what we do is, as I understand it, is to make sure that it's just like us. Is that fair?
>> Yes. The key to achieving full operating safety than a human is improving the AI software in the car. We're confident that the F4 hardware that this chip that we designed, currently made by it, is capable of achieving a safety level unsupervised, meaning if you're asleep in the car, at least two to three times that of the average driver, maybe more. And then with AI5, we think can achieve probably a 10x in safety. So these are just big deals. I mean, I think it's one very profound things that I'm saying here, and I really encourage people to go out there and try Tesla self-driving and see for yourself. You can just go to any Tesla store, they'll show it to you. You know, it's not secret.
>> Everyone here, you're benefiting if you try this and then buy it. But once you try it, you're going to buy it. So, I want to close on what I mentioned this morning on CNBC was that you're not doing this so you can get enough money to buy a beach house. You're doing this even mine.
>> Yeah, that's.
>> But you're doing this because you know Larry Page was right. You're a species that you think humans should survive. So I'm unabashedly pro-human.
>> I mean, so you're spending actually, whether you're worth another trillion dollars or 400 billion, it doesn't really matter. What are you going to do with all this money at the end? What's your plan? How do you want people to think about you?
>> Well, you know, mostly I need to have enough of ownership of the companies to be able to continue to direct their activities. But it's from a personal consumption standpoint. Um, I don't own any vacation homes. I just own sort of, you know, medium-sized house in Austin. So, and and actually tiny house at Starbase that I've worked for.
>> I've seen that house. It's tiny.
>> Yeah. Yeah. Yeah. I bought actually a few, like friends come to visit, then thought I was kidding. I'm like, "No, it's real." I bought a few $8,000 but I've done a lot with the place. Artificial turf in front, little white picket fence. So, but but like, like I said, with with AI and robotics, there will be a scroll. So it's people will actually, like, in fact, in a benign scenario, there's going to be an interesting threshold that robotics pass where it's run out of things to do for humans. It's completely satiated all human. And then I guess it'll have to start think about what to do for itself. Overall, I want to take the set of actions that expand consciousness into the future. So that the scale of consciousness grows tremendously, and that we explore other star systems, like in Star Trek, go places nobody's ever gone before, and find out if there are existing alien civilizations, or maybe there's alien civilization and we can look through their and try to what what they were like. And just generally understand the universe.
>> Where are you now? Right now?
>> I'm in, I'm at.
>> Thank you very much for today.
>> Thank you for everything you've done for me, for our shareholders, and for your manager. If you're on.
>> Thank you. What's next, guys?