Hello, everyone! We're on ToTheMoon. Technological news, Silicon Valley sites around the world. We're going out on Sunday. We're always four people. Elon Musk, he said that next year there would be a revolutionary breakthrough, and in three years, 100 percent, he said so, everything that man does will be replaced, uh, by a robot.
I wonder who and why these statements are doing. There's Altman on Channel Y Combinator telling us that we'll have AGI in a year, yes, of course.
Yeah, there's gonna be robots, and everyone else on Mars will go as he promised.
Is there any chance that his power will grow hundreds of times?
While the generative AIs start-ups were on average more, yes, well, as much as they'd do more harm than good. Because explaining how it might be useful to existing professionals, no one could.
It looks like people just eat mushrooms or something, and they're moving their fantasies into a scenario. I mean...
I mean, they should probably tell history like faunders like business, right? So what are we doing for who? I'm telling you some buckets here.
Last time we took off the graduation, it seemed so unbelievable.
So many changes have occurred in the technological world. Plus, the political agenda, that Trump became President, he was elected, and how it would affect the technological world. And this week, a huge number of people, such strong, gave interviews. There was an interview with Sam Altman, an interview with Ilona Mask, an interview, uh, Dario Amode, this is Anthropic. I've got more interviews, and I've been writing a bunch of other people on this subject. What's the matter? About what happens to artificial intelligence. And in particular, would artificial intelligence be this AGI or something super unique and replace everyone. And that's when chapter Anthropic introduced a new concept. He said that Dario Amoday preferred the term "high-pitched intellect." It's like powerful AI. And he thinks AGI is too overloaded. He claims that, uh, new concepts should be introduced. And I came in and looked at the amount of money I could use. And I have something interesting about confusion. I went to GoDaddy, and I'm supposed to be writing, "We're gonna get a bet. The domen is sold, we have to make a three hundred ninety-nine-five-eighty-two. I think that's what $400 is. I'm starting to inject, starts to inject four hundred dollars, and I'm writing, "Minimum sum for the tris ninety-nine." I understand that this comma is not a fraction of the number, not a split number, but a $399,000. Here. I've been a lot of homeworkers in AI, by the way. A year ago, there was no such thing because we buy some housework periodically for some project. It's fun what happens to prices. By the way, on this subject, yes, OpenAI bought chat.com domen.
Who doesn't know, last year, a man bought it for $15 million. Well, the man is a tough man. And then he put it in, it's a part of it. Hop-hop-hop! They're all there, so they're all converted. The amount is unknown, Elnar, yes, I think it's there?
There's up to 20 million rumors. Of course, there's no one in the exact amount, but there's rumors that about 20 million there are estimated to be...
So if someone tells you he's very smart, and he's got $2 million, then people make a living on it. And of course, you could have earned a lot of money in the last two weeks on the brick. Well, who-- let's go to a detailed discussion on what is planned for the near future.
I think there were very clear statements, and Sam Altman said that next year there would be AGI.
We said from the very beginning we're gonna go after AGI at a time when in the field you weren't allowed to say that, because that just seemed impossibly crazy.
I think it will be able to do anything that any human can do. Possibly within the next year or two. And then, how much longer than that does it take to be able to do what all humans can do combined? I think not long. Probably only, I don't know, three years from that point. So like 2029-ish, 28, something like that.
He said that next year there would be a revolutionary breakthrough, and in three years, 100 per cent, he said so clearly, everything that a man does will be replaced by a robot. And then three years later, he somehow called two thousand twenty-ninth year. Well, that's okay. People have different numbers, they're somewhere, like, three, thirty. He said that. What do you think about AGI? I think this is a very interesting subject, and it's still being discussed.
Elnar, you're very good, I think you can explain that they see everyone underneath it.
Yeah, look, first of all, I wonder who and why these statements are doing. There's Altman on Channel Y Combinator telling us that we'll have AGI in a year, yes, of course. OpenAI also said a lot, a lot of things, and then they put them behind, but they keep saying it all the time. And I wonder why he's doing this for? For investors, to maintain some image and interest in the company, anything else? Not very clear. If I actually believe that in the twenty-fifth year, OpenAI will do AGI already? Well, it's not like it's a good feeling, right? Because there is, for example, an Ilia Suzkever statement on Twitter, if I'm not mistaken, and Jan Lekuna. We're talking about Jan Lecuna, too, remember, I guess, you know, you know, I think you said, right? He's all, uh, yelling that you don't need a linear algebra to call artificial intelligence. It's just a re-creation of the matrix. Yeah? And here, as they say, that the scale, that is, the increase in neuronets, the increase in the number of GPUs and the other, is already on a platoon. I mean, when you're growing up, there's, like, ten, 100, a thousand times, the same cool effect isn't happening, right? And growth slows itself. And they say, on the contrary, that with the current approach, with the current approach to these tasks, it is likely to be difficult to reach AGI, well. In this regard, at least there's no pessimist called Suzkever. He says, like, uh, invented this approach, we'll figure out the next one and finish it, and it'll still work. But that's not the next year. And, uh, anthropics, they actually say that by the twenty-seventh year, uh, that's the I. they're doing, it's matching the level with human intelligence. And it's like there's a whole set of views, yes, that's how you, Sas, said, a week of interviews. On the one hand, the most optimist Althman says we'll do it next year. On the other hand, it's working...
Yeah, yeah, yeah.
No, he's even replaced people next year. Then he'd say, "Well, everyone, it'll be two or three years. Well, it'll be all over the twenty-ninth."
There'll be robots here, and all the others will go to Mars, like he promised.
Looks like it, yeah. And the anthropics are the twenty-seventh year, and Suzkever and Lekun say that this is the current approach to AGI, we're not gonna get there, are we? And we need to go to science and watch, there, what further approaches are. And from this point of view, if they're all put on this, and, uh, straight, what can I say? That everyone needs some points for these statements. The souqewer needs glasses that this approach will not work. Put money in science. This science company will be working on, and we will actually do new approaches. Althman shows that OpenAI is ahead of everyone, and that's what they say, uh, they're so active. Well, Elon Musk is a separate story, and anthropic tends to be careful about this sort of thing. And from this point of view, it's very interesting to see who says it.
interesting, so if you look at their interviews carefully, they're very much on the security issue, and everyone's talking about security. That we need to pay attention to security. The anthropic is saying that development will be s-- they are giving a lot of attention to Amadei interviews, saying that they will devote a lot to the episodes of cauche- quality and security. But Elon Musk says that there's a possibility of twenty percent that artificial intelligence will be out of control. But first, he says she's not zero, and then he calls 20 percent. Twenty percent, by the way, is the probability very serious, right?
Yeah, yeah, yeah.
I mean, if he's making robots and making Neuralink and the car that self-driving car, I wouldn't want to get in a car that, uh, 20 percent of the probability. I don't want to get in a car where zero, one thousand, right? I don't even want to get in a car where one's a 10 grand mistake, but twenty, two hundred people out of a thousand, like it's a serious, you know, serious.
It's a serious, well, serious possibility that artificial intelligence will be out of control, and very many things will be done and specific, yes. And, well, actually, we saw a description that the Chinese military, I think, yes, took an open Llama model, the Chinese military were there or who?
Mm-hmm.
Llama and the bases of the Llama have been taken, and have created a solution for the military industry. And, uh, it's already, obviously, yes, many things are out of control. By the way, Apple said this week he'd release a smart homie. They're all talking about AGI, and Apple says they're gonna have a smart hom in the twenty-sixth year. I think they wrote yesterday next year. Next year they'll have a device that will regulate a smart house and there'll be an I. Well, it's not interesting, is it? All about smart assistants, they've all told me about everything.
The scale is not that big.
Yeah. But at the same time, knowing Apple, you know they just don't want to be... stiff, I think, with some words. But if Sam Altman talks about the twenty-fifth year and Ilon Mac says so, and yet Darius, he says it's about to be a little soon, so they'll find a solution. Technologically, Sas, is that what you think from a technological perspective, as much as it's a real job?
Because it looks like chat.
Yeah. Well, AI is really going to give a fortune, especially in the generator code, which is in some specific direction. There, for example, I think that's what Devin Cognition does, that's the best use. There's a bad website, but it needs to be rewritten to normal fraimvors, huh, from scratch, right? But to be functional before, right? I mean, it's a de facto reflex. Because you can test carefully that the function that was before is working now. There's a big breakthrough, because AI has one of those... It's really good to wait for a breakthrough in the sense that AI will learn as a reasoning, right? Because you can always try, learn, improve, and so on. The task is specific, and the task is clearly defined as a clear task. If we're talking about models being further escalated, there's a lot of reports that Gugla and OpenAI failed to produce a larger model in terms of numbers. So I think winning--
I'll just make it clear.
Yeah, yeah, they're--
They didn't even get it, even if they started adding
power.
Yeah, yeah, yeah. I mean, there's a big break in the reasoning. That there's one more toy, which can be a little useless, in many cases useless, is true. I think there's a big breakthrough. This is GPT-2 now, that is, about one model, that's GPT-2. Again, OpenAI is probably the first to take out his O1 model, well, to the O3. So it's like a current GPT, just in terms of reasoning, right? And I think that model, she, uh, will close some tasks very well, very well. And as if AI, which will be, AI, which will improve itself, AI, which has some tasks better than, uh, almost all the industry specialists. I mean, the end of the game, yeah, that's gonna be some kind of, uh, system of agents that refact the code, right? I mean, I can easily believe that this is gonna happen in the six- to twenty-sixth year. Anyway, uh, and if it's about-- what's touched, and, uh, I think it's exactly important with websites, which can be tested quickly, websites will make them responsive, implicated, to make AI, which will be a quick test. improving AI is harder because the fedbek's onion is way longer, much longer, right? I mean, I'm gonna find out, there, and if I rewrite the model, put the practice, find out, it's not working, it just takes longer. And I think it's a way to fake it would be much more complicated. But there will be specific, light-weight tasks that AI will do before the end. I mean, AI can write--
It exists...
I mean, what AI can do, you give him your website, and he gives you an app. It's easy to believe, too, right? Because you have to take the existing function and how to move it qualitatively. That, that's a good thing to believe, too, but some serious stuff is really hard to believe. Well, as the OpenAI doesn't have a model of video recognition. Gemini has a little work in it. I think it's still a long way.
Nah, well, rewrite the website on the mobile application is not AGI, is it? And that's not it, it's not...
This is agency behavior, this is a real agent behavior that you can take a specific business task, you can hire, like, a studio, right? That's a little loud.
I mean, that's the way it is.
Yeah, but it's not, it's not a substitute... that Ilon Mac says a human replacement and a human replacement. But it's a little different, isn't it? It's just an automated performance of a certain function or, as you say, an agent behavior. And from the point of view, if you see this case about what they're saying, because they're in their statements, all of them, and Amadei, and Mask, and, uh, Sam Altman and Suzkever, they're still using them, using them, using them. That term AGI or something, they say something completely different. What's going to be-- what could be the first case on the market? In theory, what are the available ones in the market, in terms of investors? Is that what investors believe? What case? Because, uh, making mobile applications like, well, it's a good business, which, but it's obviously not a breakthrough that should cost, uh, hundreds of billions of dollars, right? I mean, it's just like a little case. Or is this a little unexplained?
I think it's, uh... I think, yes, that Anthropic has made a bigger computer-use bet. So you can run a mouse, there's a keyboard system, right? And that's how to do specific tasks. I understand that's what investors believe, but I'm not sure it's gonna happen.
You're not sure what's gonna happen-
The problem is, again, there's a lot more to be able to recognize at the code level. Mmm, everything is as simple as that and environment as automated. I'm saying there was no breakthrough at the code level. I think that's the first major breakthroughs, they'll be in the coding. The best directors in the world, they want to work on the code-generation, I think. I think it's like, uh, a system that automatically buys air tickets and plans travel, but it's not gonna be like a self-targeting AGI start-up, right? I mean, I'm talking about this, the mobile application, too. So they're working on the code, so we can see it as the first byproduct, right? There's something they made a breakthrough in the coderizoning.
And the purchase of air tickets through the interface is, I think, there's a next one, uh, byproduct.
Well, there's a lot of discussion that if, uh, with the scale of the model itself, we're going to the platoon, the other parameter that can be moved is the time that models are given to give a response, right? That's the most resonant Sasha's talking about. And right here in this direction, too, there may be some movement. If we take an agent from the site who makes an annex, we don't want him to answer as chat-bot. We're ready to wait for a number of cycles, some hydration. And that's the way the model can move. It's cheaper than either, it's a completely new approach of inventing or, accordingly, scaling up, from billions of paramet-- hundreds of billions of parameters to, uh, dozens of trillion parameters. And we have to understand that when we were in the middle of a hundred times, for example, there was a huge increase in energy consumption, we had a huge increase in the cost of learning and so on. And if we have a little improvement in the model, it's easier to stay on current models, try to finish them there, to finish them up, than to make them all mass--- to do some giant monster-like models, if that doesn't mean it's a big deal. Some of the same big increase.
But it remains very interesting in any case, but it should be brought to the attention. Lenar, I think you said really cool about why they're talking about it now. So mid-November, the presidential election was still under way. Why did everyone get so utterly uttered? Where did this package come from? Because last year, there was no such thing as saying at the end of the year, so much, right? Too open they say that. People remember that anyway, they pay attention to it. I once lied, I cheated for the second time. Well, as much as you're asking for it, right?
And now they're asking for it very actively. Let's get to the interesting topic tonight. We talked about what's going to happen with the technological world and with AI, considering the coming to Trump. If you didn't see it, look at the last edition. And, uh, today, now that we're going to move on to a new topic, what do you think from the point of view of the way out is this AGI.
That's what this arson is. What kind of case would a case come? What kind of cabs are they? You think it's from the perspective of those who look at us? Write in the notes under the video. And it's always very interesting and very interesting to the audience, to the people watching, and to analyze, and to complement us.
Write it up, keep it on the canal, you know. It's a big help to us, our development. Look, here's the question, uh, Ilona Mask. The Trump issued a few days ago, and as the appointment that Elon Musk is becoming, it is the task of managing this depart-- separate unit that will be like this outside the alleged State to monitor bureaucracy, bureaucracy of this process. But what's interesting is, in this order, uh, it's very clear, as it says, it's like a legend, yes, or how, like Trump wrote, he used a word in it about who Elon Musk is. And it says he'll run it with Vivek Ramaswamy's man. And that's, uh, interesting concept, because last time Sasha was just saying that Ilona Mask would probably be green light to show policy in terms of the technology market. And here, as an ulna Mask matched in this department, who is supposed to help clean the bureaucracy in the United States, I don't know how they're gonna do it, there's a man who basically says he's in the position that he's gonna be doing it. Big corporations and big tech are very serious, and, uh, limit freedom of speech, takes over, reset the blanket. Well, Ilon Mac, he's a representative, he's a representative of this, yes. And under all circumstances, he is also for freedom of speech, but he still has a resource that speaks of a freedom of speech. Here. And he can make some statements. People are watching him, too. How does this function change and affect the technological world? Or, on the contrary, it is only the opportunity for the next few years to move forward with bureaucracy. The Trump, by the way, has been said to be spending seven trillion dollars on bureaucratic processes, seven and seven. But it's hardly a lot of money spent on bureaucratic, bad trials, right? How does that change? Or did it just confirm that Ilon Mack got what he was promised?
I think it's part of the deal.
Yeah, it's a utopia.
But this utopia, she, this utopia, is it realistic or is it just a formality for the outside world? Just show me, look, we did it. And Elon Musk is not only a technology, but he has a role to play, but we're all gonna stay on, and everybody's gonna do their business.
Well, I think...
I think Ilon Mack doesn't do anything for, uh, just for status, I don't think he's doing anything. I mean, he's kind of exaggerating everything he's doing to make it beautiful, sold and all. But I think it means he's gonna be serious about it. I mean, it's not that. He's not-- he's not gonna be serious about this. I think he'll remove a lot of agencies and regulations that prevent his companies from developing. Here. That's what I think.
Yes, first of all, because he's the man in the neurolinke and there's a lot of regulations, because he's the man in the Self-Driving, there's a huge number of them. Regulators, as he's a man in space missions, and there's a huge number of spaceX regulations, a huge number of their own control satellites, including around the world, who can be tracked, You can't do what you want. Since he's in the social network, there's a lot of regulations, because he's creating robots, and there's a lot of regulations. Well, that's just what he's gonna do, and these questions are gonna be really clear.
Well, everybody-- all that's got to do with it, yeah.
Is there any chance that his power will grow hundreds of times?
Hundreds? Yeah. I mean, just if he can't affect any State at all now, if he doesn't have the power, well, at the Tr-- at Baden, she's gonna be, yeah, so now he's gonna be at the level with government agents, agencies, they'll be at, at, at the level of their kind.
Does that mean he's gonna be running things faster? You, Sasha, just in one of the videos, made it clear that he's the only person, Elon Musk, the only person who can launch very huge, just infrastructure projects. That's it. Does that mean that he will start them even further and influence them, including in the United States, but that is, very much in the world, to come on this agenda?
Yeah, I want to believe it because history, of course, says it's impossible. All the lu-- well, history says that people go from innovation to politics and, like they do, they don't care about innovation, but like power, yeah. And it actually seems that these are things that are connected, but in fact, not always. While Elon Musk proves otherwise, God bless that, yes, this is the only person in history who is still prioritizing innovation and business, not just, there, the existence of force and control, yes. But let's see how it's gonna be. Now he's gonna have the strength and control he's never had, and I hope it won't ruin him.
So he's gonna be a politician like this and in 20 years he won't know what Starlink is, yes, or he's still, yeah, and he's gonna make his slogans, just say another thing, right? Or he'll still be technological.
Yeah, so politics is like a tool, like a self-targeting, it's like a tool for innovation, as a means of earning money or as a means of force and control, right? If he's got politics as a tool for innovation, that's what everyone's gonna want to believe.
I guess, of all the options I've listed, that's the best option for everyone, for everyone at large.
Yeah, how much is the question in this situation, how much does he really have map blanche to do that? Well, maybe he was put in a real control policy in parallel, right? Vivek was still aiming for the Presidents of the Republican Party. So when they put a man in such an environment, maybe he is the control point. What can be done from a policy perspective, what can't?
Yeah. And I think there's just another aspect that I think Ilon Mac is an entrepreneur, and he has some things that he really cares about, where he gets passionate, aggressive and there, in general, Do, you know, how to achieve the goal in any way, at any cost, right? There, but there are also many things he doesn't care about that he doesn't care about. And that's how the State can be governed, I guess. What, like, with these, those, those agencies, I'll burn them all, actually, optimizing them, screaming their laws with the help, there, I don't know, this Grock, he'll scream all their laws and speak Groc there. made ten recommendations there. The gamble is better than almost all the politics. Well, like, I know he's gonna do this show for some of the people who piss him off the most, and the rest of him is gonna just be out and just say, "Get off twenty percent and good." And since things don't work out, it's probably a more systemic approach to politics, right? I think the systemic approach to everything is not interesting for him. I think he'll make a show. He promised to make sure it was the most entertaining show. Yeah. So he'll write, he'll write Grock's squeezing and say that I'm the one that's optimizing the state, that's, I'm the government manager. I mean, it's gonna be.
Well, he's got a name.
Replace all the people, the president, including.
AI.
There, based on the OpenAI, there's a clear picture of what's going on. So they're scared to go fight Hollywood for sure. Otherwise we'd know about it. So they're trying to say now that this is a new form of art. They're trying to invent him as they can. Well, that's what I think.
Well, I don't see any form of new art there. Real.
I think-- well, how do you explain what this is?
Like a copper art? I can't explain it because first of all, I didn't see any innovation, but secondly, it looks like people just ate mushrooms or something, and they're putting their fantasies in the script. I mean, well, even that's what you sent, it's like another format of a thing that's more like art, if it's just like a copper art and what it's generosed and okay, that's the whole concept and that's what it is. I appreciate it. Well, that's great. Well, I've seen how we've discussed it once, twice and everything. It's not going anywhere any further. What's the point of this, then, to all of us, to create such a format?
I agree. I mean, yeah. I mean, there's a lot of stories, I don't know, but, uh, I don't know, I don't know, Beckmbeth's, like, in any sales studio, there's not even a big one, there's dozens of scenarios that, you know, can't just, you know, just pretend to be in any of the world. for a number of reasons. But in general, because she always is, right? I mean, there's no match between there and there, so there's a distributor to sell.
The distributor concerned is that the quality of the actors of a certain quality, the actors of a certain quality, the director, the producer. It's hard to put that puzzle down a lot. And I know that, first of all, everyone always tries to connect with each other sometime, right? There, on the big budgets, and the little budgets, there's a lot to be lucky. And it's clear that this business, where small budgets, how lucky, well, I-- well, maybe those stories can tell, it's a test, right? The same Sony may or may any salesman have these taA-proto-testing stories somehow and, well, make some more informative decisions, what to do next. So, all these studios are sitting there, which is, yeah, there's a pro-- well ratio, one of ten projects is at best getting into the sale, which is at best. Maybe, uh, one of ten to one of the hundred. Here's AI to tell these stories, especially if they're so big, there's a choir, some sci-fi, that's something not really about real life, is it? Oh, well, or the past, where it's clear that it's just gonna be a fucking drop. Nobody wants to do that. Oh, well, I think AI's intelligent, yeah. I mean, I'm looking at it somehow. I mean, it's, uh, a way to visualize the stories that are just, you know, that's just from a big business perspective, yeah, I can't find it. So, a big business there is $30 million in stock, 30 to 6 million in stock, and with the hope of getting a $300 million fee, it's probably what everyone wants to work, everybody wants to go there. Come, in bluntly, everyone wants to work on a model where studios pay for all the risks. And the respected, respected violators, respected directors, respected actors. I think that, at the level before that, AI might well have a lot of history of vi-visualization on such good enough quality, not a stripping service, but a level of a better independent arthouse that exists now.
Well, it was a parole, like earlier, to take off a little skate, you had to buy some expensive cameras. Then there was a mass marker, and everyone started to take a huge number of things off. What you're actually describing right now is when you have the opportunity to do a lot of things, to translate the stories into a visa-visual, right? It's not just a text read, it's not a voice--
Yeah.
- to hear this story, to get her visa and support, to be visually supportive,
right?
Visual support, first date, right? Well, because as the executive is doing, they say three million tests are actually taking off something, three million tests, and you need to read it. If there's no such thing as a chain, right? The hating is like a popular comic, yes, in the previous one, or it's superpool, uh, novel, or it's already been agreed by a cool actor. If there are any of these triggers, how will the executive decide? He's gonna finance some kind of rand and then he's gonna be asked. Well, it's complicated, isn't it? I mean, it's a corporate thing, like, a logic that needs to be respected. It could be that AI, which has 100 stories, has a budget of 100,000 each. And there's a ten million budget at the studio. Well, for Sony, it's a little bit of a spike, right? Objection-- just visualize some stories for five to ten minutes, run the tests and the diapers on the basis of these tests, aaa, and make further decisions. So the executive would be more comfortable, yes, to do, because they might say, "Yeah, we have long tail, but from long tail sometimes, and we have a data-driven approach." Well, as in the ad, it's always, yeah, there's a product that MVP needs and then get some meters and see how to further develop. And then you can build this data-driven story. I mean, executives are the first thing to explain their work, their work results, so that no one can fire, no one has any questions, and so on. And I think AI here will be a big help. And, well, it's just, yeah, a lot of independent people out there, people who want their own-- their follow-up to "The Lord of the Rings," Harry Potter, uh, there, some Game of Thrones to take off. Well, that's gonna happen, too. There's something going to be easy to get, like Harry Potter. Game of Thrones, I guess, is a bit heavy. I mean.
And plus a lot of things, there are old scenarios and stories and books that people want to visualize, too, right? And they want to...
Yeah, yeah.
- to see.
Or an old IP who's been locked up long ago, to take a few of these now, but there's also when people take something old out there, it doesn't get it, it's not coming in. Again, you can re-envite-- you can visualize and do some kind of team focus. Still, Hollywood has a sense that at least 1,000 people would be fine to test.
Oh, well, they do, they have a pre-condition to be subtle, yes.
It looks like a script.
Yeah. I mean, it's--
It's like a strategy, yes.
I mean, that's what's gonna happen. So now, in principle, studios, they didn't want to do it, did they? I mean, uh, we were just dating the studios, and they all want to do it. They just can't be, they can't introduce AI. I mean, there are studios, there's a guild. Gildy says, "Nothing, you can't AI." As a studio can't say, "We'll bring AI there, for a little bit." It's kind of bad, if, I mean, it's gonna be like a process, you know, a talent job, yeah, a sales studio, that's talent. And talent says, "Let's make a new model like that, "that, uh, basically there, might be a profit share differently, and so on. A little more financial model, more like a start-up, yes, in a way, where to go through stages. And if-- it's a good thing to make, yeah. I mean, but it's gonna be, but the point is, that, yes, in general, all the generative AI start-ups in the media, they've done more harm than the good thing that's gonna happen. While the generative AIs start-ups were on average more, yes, well, as much as they'd do more harm than good. Because explain how it might be useful for professionals to be.
The truth is, everyone's watching this without knowing what's like watching some video, they know it's a little slot, and yet it's not done, they don't know what strategy.
Yeah, they understand, yeah, that's the video. Why are all these videos AI? Why, huh? I mean, like artistic quality, it's almost always missing.
Yeah.
Storytelling's missing. The reason is why even-- okay, I know it's not working well yet, but at least why people can't explain it either. And that's what I think is a problem.
Yeah, it's a negative feeling like it was like it was from the beginning that GPT chatting could be a poem, a poem, anecdote. People think I need this? But once, I came up with a poem. Well, I'm not gonna make every poem up. And that's why the man stops-- and that's the-- that's the-- that's the-- that's the-- that's the-- that's the-- that's the-- that's the-- that's the-- that's the--a-brain- from the sample, right? When you got the instrument so partially very, and it works here, I don't. You're curious today, by the way, you and I have been discussing this a lot of times. You've explained how a little other chain is full. In fact, there's a sense of strategy, why do you need it at all?
There's a whole...
Well, that's what we're just hearing, right? So that's what we hear, the matchmaiking is a long one. There are some studios that want to go to long tail, but they also have long tails. I mean, straight-up long tail, we went there doing anything, right? There's something they have in there that they understand a new audience strategically, that's where they come from, right there. Epilatum, for example, is it? Or Sony's got a business in the building, isn't it, is it? Sony's got a PlayStation, and it's where it came from that we need some sort of audience. And they're like a fucking, uh, experimental audience. That's where we need to do a long tail content, right? It's a nuance, too. It's all kind of a nuanced thing, and this technical company can't do it. They don't even think that way, do they? So no one can say that, even, like, OpenAI, Sora, Runway, that we do this kind of content. Everybody says...
Yeah, we have a strategy. So when they, when OpenAI and Sora put out six months of recurrently some kind of skating in their own-- you're watching them and you think at some point in time you think what the hell you're putting them out, right? You're gonna be doing this for ten years? What strategy do you want to set up? You want to let a man generate reels for a minute? Or do you want to give me the opportunity to generate photocopy? Or do you want to make sure you have a long movie that can be done on the scenario? So what's the strategy, huh, Sasha? It turns out nobody gives. That's a real problem, isn't it? So no one gives a normal connection, saying that if we learn to generate a normal long content, we'll take a book that ten million people read there. There's no film on this book, we can make a movie on that book, and someone's gonna be interested in seeing a whole story, maybe even with some variations, maybe some variations.
Maybe the writers are cool, right? Those who write these books, right? They have, in fact, they make, well, they're low, well, like millions, but there's one, two million a year, they're gonna make good writers, but they'd probably be out there, you know, not 100 grand, for a year. 30,000 for visualizing their books to sell even better, they could be single out, of course. Because now, well, whatever they get, their scenarios are buying, but it's not as expensive as they might want, right? So it's really good, you know, like a business, right? Unfortunately, it's not purely technology, there's this service part, because every man wants to show some kind of style in there anyway. And AI's still in some general style, and he can only be in the style, and there's animosity of the card, and so on, right? Well, AI can't figure out that the director has some sort of style that he's got some kind of filthy--
Some kind of stylist that he made some films, he took it off with a logic like that. Of course, the ch-mort will require human and small post-produced supplies to give some kind of Wyb. I can't do this yet, but to do a story that's connected, I can definitely do it. And I'm a business that is, of course, a-a-a-a-a-a-a-a-a-a-a-a-a-a-a-a-a-a-a-a-a-a-a-a-a-a-a-a-a-a-a-a-a-a-a-a-not-not-n-n-n-n-n-n-n-n-n-n-n-n-n-n-n-n-n-n-n-n-n-n-n-n-n-n-n-n-n-n-n-n-n-n-n-n-n-n-n-n-n-n-n-n-n-n-n-n-n-n-n-n-n-n-n-n-n-n-n-n-n-n-n-n-n-n-n-n-n-n-n-n-n-n-n-n-n-n-n-n-n-n-n-n- It's probably business, but it's definitely more accurate than what's happening now, yeah.
You've certainly said a very cool idea about what's on the market as a whole, if you take artificial intelligence, it's revolutionary.
Someone doesn't believe it at all, and says it's just some kind of joke that's coming up, and it's all going to go away, and nothing's gonna be interesting. In general, the market is seriously suggesting that it is like the emergence of an Internet or a computer, something very serious, but this tail that remains, what is not described, disclosed or put forward. Some kind of nonsense, or a bad decision, or someone says they're gonna do something, it's not getting to the end or it's not clear about the strategy. Ito, uh-
You know, Yuser's story is unfamiliar, but the laser is a story that doesn't understand. So everyone only talks about AGI some kind of thing, like, there, we do a genetic media, and then it'll generate itself. It's cool, I'm a generosity TikTok. Well, it's kind of cool, cool, cool, but, but, uh, but, uh,
What's going to happen, huh, what's going to happen?
But, but now more than a billion dollars in video-generation are invested in a total, a world. And the question, you mean, what specific yuser is this invested? That's what no one can answer. Under--- well, that's it, and that's okay. I mean, it's gonna be baby steps first, but there's something to do, but no one can describe a particular story-leader. I mean, what's interesting, huh? So, usually, start-ups, how are they? On the contrary, so tight SaaS, there we go, in the property of the cap, automate the invas process, and, uh, all the platform to manage the invoys in one place. Monito-- and that's what you do multitude and stuff. Usually, in the contrary, the most focused should be the start-ups moving from the safe series to a very clear understanding of the story-study phase. And there's a lot of media in this place, but it doesn't make sense. So, for example, Suno has a good case. So Suno has none of them all creativity tools that have the biggest usage. It's for the audio-based tool. So all the leading artists, musicians in Russia, know exactly what they know, and in Kazakhstan, they're now doing experiments on the Suno, and then they just overplay to avoid... Well, we're talking, that's where the leading men of the estrades are, there, wait, top five of the most popular musicians we're talking. It's happening. But, yeah, they're the ones who figured it out. But the question is, Suno, they must tell themselves. I mean, like, they, I mean, they should probably tell history like a faunder like a business, right? So what are we doing for who? So I'm telling you some buckets here, some of them are very... very confirmed, like what I'm saying now, something may not be very confirmed, but there's no way I can hear it from the creator, then There's a creator directly, yes, there's a Suno. That's why he does it specifically for who, that's what the story is? Except for the words creativity, new form of entertainment, that is something more specific, unfortunately, yes. Yeah, but it's just that I-- we met Sony yesterday, and I saw these people there, I kind of figured out what they were living. I realized that it was all for them... Well, it's like there are people who earned so much money, and it's like art for them. And there's no way that even me faster and better than you. They're like, "Well, what? Well, for us, it's kind of art, like, we're in there, like, well, books-- well, I mean, the paintings are still very expensive, right? And that's what happens, that's what people, you know, and in fact, many of these executives look at it, the art and how to cover their ass and make a lot of money. And when I was just with them, I was in the room, listening to them all day. And yes, that's how it's all going, it's getting clear. Well, that's where Runway is now investing five or six billion of estimates, fifty million revenue, five to six billion worth of dollars. Well, Sony's the guy who says that about Runway, you know, like dudes came in, showed something, but it didn't make sense. Like the dudes said, let's do an animation test. It's been five minutes since I got a chance to make a point. And the question, like, Sony's question is, what are the dudes never trying to do animate themselves? What?
Well, like, why doesn't they work if they want to, uh, I don't know, yeah, shorter, I mean.
That's a good thing, by the way, yeah. It's getting different juices because it's a technology saying, "We're doing something different." They say, "Why do we need this?"
Yeah, so it's like a tool for what? To make animation? Well, show me something's done. Well, ostensibly.
That, that's a lot in the AI toolkit. They come with some idea, like the idea is really cool. You sit there and say, "I'm not gonna be in business, 'cause that's when it's full, then I'm out. And now it's just me that's gonna cause trouble, destroy me a bunch of hemorrhoids, I'm gonna create-- why am I gonna introduce it?" And I don't-- I don't-- I don't. And I don't use a lot of tools, because, well, it's some kind of superficial story or para-- just for very smart people, that's, or for those who know. It's clear that a lot of all the workers are in the way of a number of cases. Great! Thank you so much. We're on ToTheMoon Channel once a week. Technological news, Silicon Valley sites. The commentaries are gonna be great, and what scenario would you have even filled up a movie or a book? And as you see the move, that's where-- what's going to happen here, ahead, and what kind of decisions do you see that you like more? Maybe someone likes... You're here, Tanya, saying Sora released some hallucinations. Maybe people like it.
That's enough.
Write-- yes, that's good. Write what you like. And keep the channel up, sign up, keep your signature, and keep your video and your liquors. Thank you very much, everybody. I'll see you in a week.