They have a whole cemetery of projects they started, announcing, which is very good as the users, but eventually they disappeared.
People are given, uh, awards, ordens, there, titoons in the UK.
Well, with everything that everyone understands that it's a serious threat, it means everyone's talking about a dozen billion dollars, that it's super-high-threat.
When Sam Altman comes in, for example, and says he's gonna raise $7 trillion. It's not billions, it's trillion dollars.
Which, too, will be a true emotion. I think it's utopia.
How many budgets are those of which countries? Hello, everybody! We're starting a completely new project, technological news. Although we broadcast them from the Silicon Valley, at least three of us are in the Silicon Valley all the time. We were gathering news, and I was drawing attention to the fact that three or four news were very connected to the creation of avatars or people like AI. The whole world is talking about indefinitely, that... that... that avatars will replace people and people don't have to act.
We're all four real men. And what's interesting is that we're-- uh, that's, uh, that business incubator, uh, probably the largest, top-down in the world, uh, and the coolest in the United States is Y Combinator. And in Y Combinator, uh, there were 215 companies out there. We've been looking at them in, uh, Wednesday four, and it was literally just a couple of days ago. And there were two companies that were represented, uh, through the avatars. So the company was not really drunk, but the avatars were drinking. And I was wondering, uh, right after I was drinking a girl, I was watching a little background this time, and I turned around and told a man, "Look, is that not a real, uh, man? Like avatar says. And there was a whole discussion with the investment fund on the topic: it's still real or not. I mean, it was really hard to tell. And I want to start, uh, from, uh, with the news, that's what's in the avatars. Ah, we'll start with a very simple start. Uh, Google's got the taxi. He actually sent a relatively good release. It's a demo of what you can create, uh, avatars, and we can broadcast through these avatars. A lot of different companies do. And I've got a release, uh-oh, well, I know if Google lets it out, it's gonna be amazing. Uh, what do you have in mind about this? So, Sasha, you're like a man on the front lines, especially in a top-top left in AI, and especially in the video-building, what do you think about a google-bag?
Yeah, Google does-- you have to understand Google that Go-- what is Google Research, and he does, uh, actually, they have enough resources to resurrect in every direction, and in every direction, they have enough resources to do resurrections. produce results near-state of the art. And the burglar is one of the examples. I assume that it is practically impossible for them to put this technology or to put it into the product because of the ends of the di-dipphei. It's now, uh, the year that the president is elected. And of course, everyone has been lewd, uh, sensitive to this subject is very high. Ah, I didn't see anything new in terms of quality for myself. I'd say that HeyGen was in the animation already, full of characters. I mean, not only so the avatar can be a little bit, there, whacking hands or turning a little head, they're already going to a full character animation. Ia can see that there's a demand for this kind of technology. The two biggest companies in this area are Synthesia and HeyGen. They both have more than 20 million revenue per year. Which, of course, highlights that these technologies are in demand.
Yeah, about Google, I'd add that, Sasha, you have a little too much optimism about that. Ah, they have a whole cemetery of projects they started, announcing, which is very good as the users, but eventually they disappeared. Yeah? And that's why the burglar might be the same. It's a good resurrection box of technology. They show the level at which they can actually meet the challenges. They may use this for some of their projects, but it is not a fact that it is some form of release we can all use in the
near future.
Oh, I wonder how, uh, a rash-and-tie, why this demand, uh, is this a replay of someone else's voice. I mean, there's one of the news, I don't know if we're gonna discuss it or not, about what, uh, people are getting, uh, rewards, ordains, there, titoula in Britain for their development in this field. Why is it in the voting field, not in the field, like, a visual or something closer to me? There I can understand that, uh, translation difficulties or, uh, there, just in my opinion, that's the ratio of how much people can actually, uh, be frauds, use it in some of their own. The threat is much higher than the benefits that can be derived from such technologies.
But that's right, my pure thinking on the subject.
It's really fucking, uh-oh-oh! And a very cool thought, uh, what people will be doing with a focus. And from the viewpoint of the visual avatar, it's infinitely all being discussed as some and-- and on the one hand, a drowning model to show what you're looking at, so we'll replace all people. All the time, they're talking about what they're supposed to put, uh, put people in. And here's about Britain, what you sounded like. In Britain, uh, Demis Hassanis is a chapter, uh, and one of the founders of DeepMind is a knightly title. 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-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-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-a-a-a-a-a-a-a-a-a-a-a-a-a-a-a-a-a-a-a-a- And Jan Hogart, Chairman of the Institute for the Safety of Arts, founder of the Statap Sonnic, acquitted... There's a CBE, that's Commander of the Order of the British Empire. And the same award was awarded to Matt Clifford, an intellectuals advisor to the British Government. I mean, obviously, it would seem like we all saw in films and on the news everywhere how hard it was to get, uh, this league and the super-crunched British community. And here, people who do artificial intelligence, they become even perfect, incredible. Just like it was in the framework, uh... I think it was very much like that when, um, the cryp-crypt market was built. The new people who started to be present in this market have become so new in the world, er, and for many countries like Idols or the government wants to touch them, too, and beside them, I'm not gonna lose it. So power doesn't want to lose the... uh, access to artificial intelligence. That's very much, of course. About the voice. Let's talk about it, uh-oh, voice history. You said it was a voice. This week, much discussion was held that the OpenAI released, uh, more accurately, the appearance of, uh, the modules and the operation of the voice generation, which can be used in all different directions. And it also showed that it was new that, uh, uh, access to the model had, uh, some companies in advance, including HiGen. Sasha, you've mentioned HiGen now. And, uh, in terms of voice generation, that's a very interesting subject. Again, now closed in the States, and OpenAI officially stated that I'm not-- we're not gonna open it because there's gonna be elections in the States. What's going to happen to the voice? Can this even be recognized by some mechanisms, a toolkit?
What's going to happen to the voice?
From the standpoint of the vote, of course, it seems that this direction, from-- well, I think from a scientific and technical point of view, is a direction that will be decided first. Uh, there are two aspects. I mean, this is probably the simplest thing to do. It's a-- it's got a direct application-- that's a direct application-- that's a sound of books. So, today, it was literally the news that New York Times could be listening to all the news in New York Times, yes, uh, as far as the AI sounded. A-a-a-a-and-and-a-h-- then also reading books by a non-professional actor and a duplicate not by a professional actor, but by a duplicate by AI. But unfortunately, it offers very many opportunities for frauds. I guess there's a little more to look at how technology evolved. And OpenAI has this technology called Whisper. She's the one who's got to be the first person to be involved.
She's-- she's still, she's the first to recognize the speech. But what I wanted to say is that, despite not being a technology, it was not a key technology for the OpenAI, it was still a state of the art. And one of the big questions for me is how much OpenAI will invest? Because, on the one hand, it's low-hanging fruit, on the other hand, it doesn't approach AGI. How will they balance? It is interesting because, in many ways, I think progress in this area is determined by the extent, in principle, of progress in AI, the extent to which OpenAI is doing so. Again, there are only two such leading companies that do a lot of speeches, that's OpenAI and Google. But Google does not usually put technology in open access, but only open-AI. And that's why I have the main question.
The news about Hassabidis is that it is not now Demis Hasabidis, but Sir Demis Hasabidise after he was, like, e-e-e-court.
I was just going to add to the examples Sasha had said about the sound. There are interesting starts. I don't know what stage they're at, but they're definitely, uh, they're changing as a result. Uh, it's a story for kids, a story for kids with a parent voice. Yeah, when you choose a story when you're putting your voice down, and then your voice is reading that story to your kids. Yeah, and there's a lot of different ideas, so, say, there's a lot of positive use scenarios, uh, voice generation.
Uh, not a horse-- I can be oriented in ranks and names and in the right utterance. Ilnar, therefore, I didn't say "sir." And with the kids, I'm saying that I have a friend of mine who says, "How cool I'm gonna drive in the car in a few years, and I'm gonna have a lot of avatars to do. Different jobs, and one of the avatars will be communicating with my child. " And by the word of the avatar, and by the words of reading, uh, fairy tales to the kids, uh, with a cool voice-and-dance tool, right? It's a tragedy, because so many people can't say-- spend time with the kids. I have four kids, and my wife is 13 years old each night reading a book. I look at her and I think, "This is the power! It is. " I'm so hard to get over it, so I can read it. And it's emerging with new technologies, it can do as well as, like, there, I don't know, support people with a speech that's been broken, there, translations into different languages, uh, like, reading New York Times. Any language, yes, in Chinese, who you want to, in Russian, in... That's super cool stuff. At the same time, of course, and severely restricting, very limited space. What else did I want to ask? From a technology perspective, uh, how hard is it, uh, how hard is it to create? Nasco-- if, for example, Google doesn't, uh, open access, OpenAI is making some part of the open access. How many new companies can be created and truly create and open access, uh, where people can use it, you know, from the point of view of faeces and big problems in the whole world. The world?
That's a very good question. And, uh, right here, well, there's got to be another way of knowing, yeah, that's a point-- well, like a transfer, roughly, it's called it. So, it's when there's a teacher, that's some superpowered network, right? And you can learn from her exits. I mean, practice shows, I think that there was one, like Stanford's work on this, and they showed that literally enough, uh, a few dozen thousand examples, from, in abruptly, output to ChatGPT to show quality comparable, right? I think that's a big deal, because, uh, OpenAI is likely to add--- will, expand, the possibilities of its model to generate and pictures, and videos, and voice. And the villains, they can do, in a rude way, 50, 100, 500,000, a million requests to API Open and teach their model and not tell anyone, right? And we can't even get away with this...
and we can't even trace, uh, who has access to such dual-use technologies, if that's what we can say.
At the same time, there's a big problem. I often bring an example. There, I may not be over there straight, you know, I'm a big deal of a way to develop and know how to create, uh, a story like ChatGPT or GPT-4, right? But I understand there is now a huge struggle for who owns chip production companies and where these companies are. I often repeat that, for example, in Russia, it is not possible to create GPT-4 at this point because they simply have no such access to equipment. Because in the States, uh, there is the largest producer, NVIDIA, and NVIDIA, uh, the transfer of all NVIDIA capacities is generally managed at the State level, at the government level. And there is, uh, two news this week, that S-S-S-S-A was asking allies to impose additional restrictions on the operation of microschem equipment in China. The Baden Administration is trying to further prevent Beijing ' s ambitions to create advanced semiconductors. Because we never understand, and what's really going on in China. And then the second story, which means Google, Intel and Qualcomm are going to unite to create a NVIDIA competitor, which will be for the company, well, given that everyone knows that it's a serious threat. So everyone's talking about a dozen billion dollars worth of money, that's super-dood. Uh, wouldn't there be something that, uh, who finally has access to these technologies? So on the one hand, you're saying that someone might connect to the technology already in place, like OpenAI. OpenAI, like, uh, conditionally, an independent company buys and has access to, the technology bridge--savingsers that are distributed by the US government.
How does that chain, how long it can be, and who's regulating it?
I guess I should start with NVIDIA, she's kind of designing cards, and they're mostly in Taiwan-- Taiwan, that's, like, TSMC, right? And that's why I think it adds to the additional pictance of the situation, which, as if those cards were produced, in a rough way, hundreds of kilometres from, uh, like continental China, right? And they, uh, on the de-- well, distribution is the main one in the U.S. dates, right? It's so-so-so-so-so-so. And there's a special feature. Of course she's worried, I'm guessing, China. And, uh, I'm-- I'm getting it, the only thing that we need to know is that China has its strong enough Huawei Ascend, uh, all the 900 series, and they're powerful enough. I understand they're quite comparable to A100 in quality, speed and maybe even comparable to a s-- with a 100, uh, NVIDIA card, which is now the most popular in the world. Mass-Market. Uh, I guess I should, uh, say something about, uh, what I hear from some big fund talk, uh, what's for, like, countries from, uh, Saudi Arabia, uh, actually, processors are one of three priorities. So, there's a priority, it's energy, food and people, like, right? It's the most key. And indeed, tranches are a comparable priority, that is, like the fourth important priority now. So when I know that many people have a lot of skeps when, for example, Sam Altman comes and says he's gonna raise $7 trillion, not billions, and trillions to his initiative with a few million dollars. production, like these processor factories around the world. On the one hand, it seems that it's a lot of money. On the other hand, this solves real problems for a large number of countries in terms of independence. That's why it doesn't seem like it, doesn't seem so impossible. And I think that while America has control over the top chips, they'll, uh, have some kind of leverage, they'll use them as much as possible.
I think that whoever's thinking about the $7 trillion to set up a chip company is a different level. And that's what everyone needs to hear. That, that's right, that says, right away, that everyone has an answer: can they be near them? And Ia is a lot of people who tell you what's doing the same thing. It's a thing to do. The money itself, the money's... it's a little bit of a shock, it's just a shock, right? Thato, that's not real.
It's three and a half of NVIDIA capitalization, right? She's about two trillions now, isn't she?
How much is it?
Yes, 2.35. How many budgets are those of which countries? There are many budgets of the Republic of Belarus. I'm like a Belarusian, so it's a lot of budgets, really. I'm even-- 140 budget, um, yeah, 100 s-- if that's the way it is, I don't know what a plus-minus is, like, more than 150, probably a budget. Here. It's, it's a lot, yes, it's very serious, it's very serious and serious... I guess that never happened.
I guess that never happened. Did you ever discuss things like that? Did someone come in and say, "I want to go to a technology project, ", even factories and hardware $7 trillion"? Was it even?
I-I'm probably the youngest, I can't know that, yes, but I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I- But, uh, I'm guessing that's a demand, because now it's so hard to be that it's the currency of the future, yes. And Sam always talks about it. And I think he's been able to create this narrate. If it's about keeping sovereignty, a few big countries, it might be worth it, I think it might be very much, I think it might cost a lot, of course, very expensive. I think that with-- I understand, Sam has already clarified that $7 trillion is like such investment for years. I mean, not $7 trillion at once, but I think it might be half-semi-semi-sex-supposed. I think it might be quite possible, and it's possible that, uh, that's how the company would be built.
Tanya, that's what you're saying, that now it's all about the voice, and that little one about the visa is still some cognitive distortions, right? Ah, just us, there was a lot of voice news on this and last week, and it seems that, uh, there, everything about him. And actually, scientists just dig up different sides and now they're getting to the vote, yeah. If you remember, there was a parole, um, a year and a half ago, We got the votes, yes. It's just going through some of its own. It's just that it's his turn. It doesn't mean they're the only ones who do it. Here.
But this week, there was a lot of news about him.
Yeah, by the way, as you keep thinking, Ilnar, I wonder what Sasha said, uh, about reading books to the kids, and what, uh, I think, is in all this artificial intelligence, which is basically, I can't be wrong, though I can be wrong, it's, uh, emotions and attention, yeah. Which, for the kids, is where for the couples, for friends, uh, most important. And I wonder if, uh, if it's possible to create a quatar that, too, will emotion be right to pass and, I don't know, to children love, care and so on. I think it's utopia. Here. So, uh, reading books to the kids, you know, but if you're gonna invest in that crazy money and, uh, awards, I'm gonna be wondering. I mean, I think that's why there's something, uh, more serious targets than just audio books or, I don't know, the creation of avatars to be a multi-functional.
I don't answer the possibility, but I can only comment on what you see, we, uh, audio books, they won't replace a living man, but you just have the possibility that this voice isn't the voice that's the voice that's on the side that's that. Audio recording, yes, it could be your voice. Mm-hmm, I'm, like, well, it might be a little unfortunate example, but when I was a student, I was very popular with the Night Watch, and I really liked the, uh, audio book that Habenski was saying. He was just filming one of the main roles and, accordingly, he reads the book himself. And it was a great pleasure listening to his voice. I see that there's probably, uh, voice generation down there, professional actors. But somehow, if everything goes on, why not? Mm-hmm, at least in the audio book, it might be quite interesting.
Now, actually, it's pretty good enough to make the same Haben or any actor's voice. The models already work pretty good in the
generation.
The ad came to the head that, like, listen to books, I don't think, but listen to the tape as Snoop Dog's voice. I don't know why I'm the one who's the one who's making the Snoop Dog, but they've made a whole application on this subject, I think, that he'll read these texts.
I think the subject is in love and emotion, it's definitely drowning because we're talking to you, and we know we're real people. Ia, uh, you see each other every time and you want to see each other. So Sasha and I had lunch together last night in Los Altosia, uh, great time, yeah. Or, Tanya, we're gonna have to-- I'll be at your place on Sunday, there with a family at the bottom of your son's birthday. And you know you're dating. So it's a different feeling about this heat and meeting. So, of course, it's so different, but it's very interesting, and in terms of, uh, technology and voice. The voice certainly offers a great number of opportunities, including additional, uh, development and communication, in cooperation. I mean, I lived a year ago, a little bit of time, uh, in Bali and a few months, literally, and there was one American. He told me he had a girl, uh, Chinese, and China, and China, and China, and he had a device in China. You wear these headphones to him and her, and they're totally living together, and they know each other perfectly, totally perfect. I mean, real time is moving everything. He said he was interested in making a comment that the product was not being converted into the U.S. market categorically. The US is running around with these products because they change, change the market at all and what happens, and destroy a large number of old, there, existing businesses and everything. And still, actually, this is a hardware super-right worker, well, a little bit of it. It certainly appears to be eating, but it's a little bit. And you know, of course, what a great comfort. I mean, if I came to some conference, for example, and at this conference I could wear light ears in real time, build on, uh, this man's transfer, with his sense, with his senses, It's the intonation, with his own temper, with all the others, right?
That, of course, uh, well, that increases from-- that changes the full potential. Come back, let's get back to the news about Y Combinator. The important news is, in general, the technological news in the world, because yet Y Combinator is still the most successful launcher in the world. And I was looking at all the companies in Y Combinator, just over two hundred, uh, fall or when it was in the summer, I think it was last September. And now it was this week., 100 seventy companies out of 200, 40, not two hundred and fifteen, two hundred and forty companies, they're AI. I mean, the whole AI trend. A huge number, uh, companies got investment before they got out, uh, their output. Uh, what is this, what kind of company? A lot this year, now, in this set of companies that are created by faunders completely without experience. So people have no experience with the start-ups. They came and got into Y Combinator, went to Y Combinator for three months, and their companies estimate $20-30 million. They have nothing, and all AI, just AI, right? And Y Combinator, Y Combinator, continues to make crazy Xias, you know, crazy X. I think the last, uh, last fund of 14 X, 14 X was in Y Combinator. AI, all, all about AI. And, uh, interesting subject. I'm building a Davidov and Sasha foundation in the Davydov Foundation, right? Investors, as investors here in AI, and he wrote this week that before he said I was investing in AI, it was something separate, and everybody looked very strange. Now that you're investing in AI, you're kind of investing in a computer. It's like a word, isn't it? Computer, Internet, all that word is used in every single place. Everything. What happens to these companies now in these accelers?
New technologies will be available, will they be more or will begin, er, some-- lack of progress in terms of new interesting companies?
Yeah, I'll probably take the word fast, and I'll say that first, uh, uh, you know what-- if you see-- what you need to know now that Y Combinator is raising two billion. A-a-a-a-a-a-a-a-a-a-a-a-a-a-a-a-a-a-a-a-a-a-billion, I-a-a-a-a-a-a-a-a-billion, I-a-a-a-a-a-a-a-a-a-a-b-b-b-b-b-b-b-b-b-b-b-b-b-b-b-b-b-b-b-b-b-b-b-b-b-b-b-b-b-b-b-b-b-b-b-b-b-b-b-b-b-b-b-b-b-b-b-b-b-b-b-b-b-b-b-b-b-b-b-b-b-b-b-b-b-b-b-b-b-b-b-b-b-b-b-b-b-b-b-b-b-b-b- So they have far-reaching plans, uh-oh, and all of their content is AI. Even if you're looking at them YouTube, they've got all the AI content, they're building a brand around it, and it's very common, which is the nacho--- that's in the Valley, that, uh-a-a--- the-- the last ten years--- it's been gone. You've got a hundred billion company, right? It's already been done in Generative AI, two of them, two of them have been established. It's OpenAI and Anthropic will be the same. Iia is more likely to be a few more companies established by 100 billion. So for the hygaphones, like Andreessen, Coatue and others, for them now, they have the opportunity to invest in companies they haven't had a lot of years before, right? Who hasn't been in the last ten years. But if you look at them, they're all rab--
They're not gonna get the too for them, but, uh, how much is it to go to exactly what kind of industry? So it's gonna be a start-up that's connected to 3D, like, uh, interior design or exterior design. It's gonna be the start-ups that gene-e-e-e-e-sing the voice, it's gonna be the start-ups that generate videos or something. They're very hard to make a specific bet. And in this regard, I think that Y Combinator is judging, like-- it's like ETF in a way. I mean, because there are several hundred start-ups that AI do, they cover all the lines from , ad-tech to defense tech.
And so, c-ca-e-e-e-e-e-e-as if Y Combinator's investment is actually, uh, ETF on AI start-ups.
Look, I'm curious, now, Anthropic, uh, and we talked about you yesterday, and this week, we've been right around the big-- well, uh, uh, uh, uh, uh, in March, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really, really Her-- that Anthropic models are more cool than GPT-4. Not a beef-- plus, but the OpenAI, however, this week says that it will open up access to GPT chat without registration. Would it not be like all companies, including those that were very closed, that did not want to do as O-O-OAIpen, would have to open up a lot of things, and that would accelerate the whole The opportunities and speed up the market.
Oh, I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I- I mean, uh, in the case of Google, I think, for example, they can win the market. So, neither-- Google and Microsoft have a competitive advantage that they can, uh, act like, uh, insurance company to, uh, in case of legal problems, they, they're, uh, Insurance as a consumer, yes, of, uh, potential litigation and court risk. I mean, I think that when I think about how big companies can act even more aggressively, I think about it first. Ah, and I think Anthropic has them now, a-aah, golden hour. I mean, they've been company for a lot more than two years, and they've had a model for the first time that's better than... What's the OpenAI? And I think it's in some sense their moment to take over the market. I looked at the analyst on Semrush, on their Claude.AI. Ah, there's a growth in the number of users, but he's not crazy. I think every company has a star hour, a pick. How will they be? Uh, I think it's all supposed to be... Sam said that GPT-5 would be as good as GPT-4 as GPT-4 as GPT-3. So he actually promises a huge badge-- oh-ho-ho-ho-ho-ho-ho-ho-great jump as a slide. And that would allow for the overpass, including the last Anthropic model. I assume that GPT-5 release may occur at the end of this year.
A-a-a-a-and--and what tells us that Anthropic, probably has at least three more months, maybe six months.
Well, the news was Amazon, uh, uh, was running his options. So they had a deal last autumn, and they've been paying money, which is literally, if I remember correctly, late March. Ah, so at least Amazon believes in Anthropic, and he thinks that their peak, well, if it's even up, will last for a while, and maybe they can get up.
Yeah, the pick must have been a-- a pick, probably not yet at anyone's. There's not enough people who know what's going to happen, and what's the peak there.
Yeah, I think that's a... that's a very good point. I'd like to get a little loaded here because I'm working with Google, and, uh, actually Google, Amazon and Microsoft, they're-- they're so sympathetic at the moment that they're playing at all levels. They're actually reproducing video cards and, uh, re-selling models. They're being re-selled as video cards to train models to fine-tune models to start models. I mean, they play at all levels., if AI is really gonna be a big topic that would lead to a new type of economy, right? I mean, a new type of economy, when it's not eight hours just to googling, to communicate with people, and that's all gonna be automated. And there's something you can do in eight hours, yes, the computer's work can be done in an hour. If AI is really us, if he really is going to move us forward to this future, then O-- that's Microsoft, like Microsoft, Google, Anthropic... Sori, Microsoft, Google and Amazon are definitely gonna-- will be beneficiaries. Ianthropic nacho- and, and Amazon was the chief leader, leader of this field, a-a-a-a-a-a-a-a-a-a-com computer service, in terms of cloud services. And now, I think they're in a big risk of getting back because they don't have a strong model. Google has Gemini, who's the best model in the video-to-text. OpenAI has a number of powerful models, yes, we talked about the voice and text, and about the video, and the pictures. And if you're talking about Anthropic, they don't have a strong model.
For them, investments in Anthropic in this regard are of great importance, of great weight.
Now, let's, uh, an important story that, uh-a-a, is debated and published a lot, and, uh, um, with an account of, uh, a-a-m, learning experience, 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-a-a-a-a-a-a-a-a-a-a-a-a-a-a-a-the-the-the-the-the-the-the-the-the-the-the-the-the-the-the-the-the-the-the-the-the-the-the-the-the-the-the-the-the-the-the-the-the-the-the-the-the-the-the-the-the-the-the-the-the-the-the-the-the-the-the-the-the-the-the-the-the-the-the-the-the-the-the-the-the-the-the-the-the-the-the-the-the-the-the-the-the- Ah, there's, uh, Devin, yeah? And, uh, there's a lot of things around this, that there's no need to-- and that's a lot of them, including the Y Combinator launches, and the last set, and that's, uh-a, that's not what the programmers need to do. I'm sorry to have to make a code. Oh, and I want to talk about Devin's little news tonight. Iia, uh, what is this, what's it gonna take, and what's going on with it?
Well, the news is, actually, yeah, it's not new, but we haven't discussed it yet, so let's touch it. But there's a few things here, huh? That's the first thing I heard about Devina, I went to the designer's website and went into the Vacancy section. And on the one hand, they're offering, for advertising purposes, that Devina can hire herself, uh, for some purpose, and on the other hand, they have a job in the vans that they need the software engineers. Here. There are some jokes on this subject, by the way, in terms of what we're showing and what we can believe, right? They say that Devin can be quite a number of low-paid programmers who are responsible for this task. I don't know how much it's worth discussing this, but again, we've only seen videos on which we're showing some movement, right? That the terminal is starting, that there's documentation here, there's something else here, but as if there's a hood under there for now, well, at least I haven't had the chance to feel, has I? If I'm there from ChatGPT, I can send a request and at some speed, get an answer, so Devin and I haven't had this kind of interaction. Here. And by the way, after that, literally a week or two, it's a open source of analogs. Again, it's hard to compare them. I think we should start, try to touch them, who also position themselves as, uh-a, AI programmers who can do some sort of thing. There must be a little more Sash here to say. Still, in the Silicon Valley and c, uh, directly related to software engineering, too.
That's why I'm gonna be interested in hearing you, too.
Yeah, first of all, I was gonna say that I have enough to-- some comments on this, but I have to-- but I have to admit that, of all the launches in the first quarter for the early stage of the start-ups, it's the most important thing. successful launch with all, all, all, all the start-ups. So they're in the first-- that's how the successful launch metre is, for example, the number of viewings on the tweet. They have 30 million at the moment, that's a lot. Oh, and it's clear that they, well, because they're only got 20 million fandings, they couldn't buy all the influensers there, right? So, people are really talking about this and are actually discussing it now that they will raise the next round of $2 billion. Which is likely to have actually increased their estimate 20 times or so. Now, a few comments on this. I worked a little bit at the Yandex and I worked at Steam Chat, and I can tell you exactly.
Snapchat. And I can tell you, if we look at American companies, the too is a no-no-no-no-no-no-no-no-no-no-no-no-no-no-no-no-no-no-no-no-no-no-no-no-no-no-no-no-no-no-no-no-no-no-no-no-no. Oh, well, uh, we-- you probably heard about the OpenAI payroll for two million, Meta for 1 to 2 million, yeah, the developers. And that's a huge salary, and they don't pay for a man writing a lot of code, do they? And not for saying that a man writes a lot of codes from scratch. The leading developers in Snapchat who earned comparable amounts, I think, were mostly professionals in the way of dealing with complex business logic, especially in coins. Because there are many different coins, uh, steakhulders. Business logic becomes very, very difficult, and, uh, skilled to understand the B-- understand and business logic, and understand the complex code just, uh-e, and results in such high compensation for work. I think that, um, we don't have any information to assess the quality of Devin's work, uh, with the big code bases at this point. What many start-ups do, I think, is the most popular thing, a start-up Magic. They say different. They say we're not writing the code from scratch, and we can remember the whole code, that is, in the context window, that is, in bluntly speaking, we can put the whole code--- the whole code base, and then, in blatantly, the model. It can edit this code base, rewrite, supplement, improve, and so on, find vulnerabilities because it can remember, yes, to itself, the entire code base. It's so hard, again, in the case of Magic, it's not clear, it works or not.
But I think there's an application enterprise that's more understandable, uh, a fairy tale they sell.
Devin's here, more like, he's, uh, about the other, probably the level, right? I want to create a website, an audit site, make some simple web application and some such home-grown tasks. Maybe that's what he's gonna do, right? But it's clearly too early to look at such systems.
Yeah. And so, I mean, I would be c-- I would have talked about Devin, yes, it's in the con-- in the context of no-code tool. So, here's the thing, and now, and then, the question for investors is, what do they think about the no-code development platforms and how much should they cost? I mean, uh, general, it's a huge trend. I'm not sure what he's-- I'm not sure that the 2 billion evaluation is, in no-code, it's applicable because it's important that no-code can still cover a large number of businesses, yes, and a large number of businesses. People. So very important, important, the product is important, as much as UI is, the product itself is important. Since this is not an enterprise-building decision where a AI can be connected, which is the decision for a wide range of people to create a site, uh, a vi-twitness site, I think it's necessary, uh, More food meters are required to justify such an assessment.
There's also a story that the emergence, uh, of these, uh, functionality and such instruments, it actually expands the market very much. For today, a large number of people cannot work with programmers. It's just impossible. They have no access to them or no access to them, or they don't understand how to interact with them. And the appearance, uh, so lighter, light point of entry, light-opening, uh-e-e-e-h, might even expand the use of technology for the larger numbers of people. On the other hand, of course, the Haip is very much, and the people are telling so many fairy tales, uh, that it's just impossible to... listen to a very practically lu-- very many conferences. People, um, whatever they say. I've been at one AI conference recently, and there's a man--- telling me that there's something that can be done in the sales. And once you ask the question of the reality of this, the real boxes, the real business, the mass production, the mass use, the answer is as if it were gone.
Of course, there are many different tools, a large--- well-functioning, well-trained, but spaceships, uh, until they're gone. We'll watch. But the subject itself is, of course, very alive, very interesting. Tanya, I think the first programmer in his life will not be real, but definitely AI.
I don't know if I'm gonna need a real or unrealistic software program, but I mean, you've narrowed it down to the programming area, but it's actually like our reality, there, I think, years after. Five to ten in all spheres. It's from a series of like, Sas, there's a robot vacuum, yes, that replaced us, uh, cleaning agents, yeah, people, there, with little, low-skilled work. I think there's a whole lot of things where there's a formula ready, yes, be it code or, like, a bunch of ready formulas, including design, fashion, anything. And that's what's gonna be replaced first, which gives access to a large number of people, like, getting an interior design, right? He's gonna be a little bit of a high-level, a little bit of art, right? But it's gonna be much better than what they do themselves or, there, with designers that are low-skilled. So I think that's gonna be everywhere and everywhere, uh, not just programming, but also in our daily lives. Well, websites, cards, that's just what's on the surface. I used this great thing here, by the way. I don't know how much it is, uh, well, the program you just rent a room on camera, and she makes you drawing blueprints. Not only the top-view, but the side dimensions of all the furniture are still good. Well, I mean, it probably wasn't a 100% precision, but, you know, I'd love to get rid of this kind of job in my field. It's got to take measurements and stuff.
That's what I'd love to give to some artificial-- some artificial intelligence, a programmer, and so on.
Speaking of the robot plumbing, that's the news we're gonna finish. I bought the most modern robot vacuum last week. And so, since I have a lot of windows, I just don't know how many, hundreds. And this window was the first question I had to ask, "Is he jumping between the windows himself?" I said, "Yeah, by my hand or my wife's hand?" And I'm supposed to be my friend today, uh, sending, uh, a robot that, uh, walks between the floors. So he's like that. They made a robot vacuum, standard, and he's on the stairs, so he's getting a little bit like a dog, so he jumps. So, but it all looks like a kindergarten anyway. I'm a friend saying, "so you can order a robot, it's worth $3,000, $30, it can start to learn windows to wash." I say, "I understand correctly that when I learn, he'll break a window?" He says, "Well, I guess." Well, uh, today is our first edition of the Technological News. We're four men. Cream valley, three, silicon valley, three, silicon valley. We meet once a week, exactly a week, at the same time on YouTube. We're meeting you on YouTube. We'll definitely expand on other social media. We have a great team! You have to sign, uh, put the liki on, so it's really helping the channel move forward. Especially we have a new channel, a new team. We need support, we need support. Any comments, of course, all the comments, they're moving the slots forward, so any comments are written whatever they want. I liked it, I didn't like who liked it more, like a girl in blue, a guy in a jacket or a guy with a lot of books behind him. Either me or me. See you later. We'll meet in exactly a week. Thank you so much.