Plus, there's a parallel history where Apple says, like, that there's no way we're gonna release AI. Or Apple said he had postponed the idea of integrating Meta's artificial intelligence from concerns about confidentiality.
Unfortunately, the news has become more expensive than real change.
It's a global trend in the struggle for attention. We'll get some attention, and we'll figure it out, we'll show you something. But people like that have a focus on a certain company as clients.
Very interesting. You have a point of view that is completely different from mine.
What the hell? Say it, professionally, recheck the data. You're telling me nothing. Half of me turns around and tells me this voice: "Sasha, can we talk like that?" Hello, everybody! We're on ToTheMoon.
We're always four men: Elnar, Tanya, Sasha and me, Sasha. We're filming technological sites, um, discussing interesting news from the world, the US and the Silicon Valley. Oh, look, I want to start. Now, we'll go to OpenAI, taking into account, uh, a lot of, um, different things that they've got. But I want to raise this subject at the beginning, which is the last quarter, as you can see, some companies make statements. So now everyone is trying to make statements, taking into account, uh, the company's cost. There, uh, we see NVIDIA, there, and it's, uh, it's, like, a little bit of a time, and, in general, there, Microsoft has returned, and it's got a big back, up there, compared to Apple. They make statements. OpenAI made a statement that they're gonna release, uh, this very serious voice, I don't know, assistant. It's not a voice-assistant, yes, there, a voice that can be spoken in real time. And Apple showed a lot of extra stuff at the presentation that's gonna be there. And in parallel, the OpenAI said we'd be up there a little bit. Apple, at least, I have a few bets. There's actually very simple things. AI's not even there, no macOS or iPhone. At least I didn't see it. Maybe there's something coming up at the time of the release or someone's. Plus, there's a parallel history where Apple says, like, that there's no way we're gonna release AI. Or Apple said he had postponed the idea of integrating Meta's artificial intelligence from concerns about confidentiality. Although we don't know at all, but in-- iPhone, ChatGPT will be available or not. Here. I'm not against statements that don't come true or that come out later, but, um, probably the last one, at least the quarter, I think they're too much. I mean, there's too many of these gunshots. And if I stopped taking these shots at all and looking at AI, that's when many people keep making different statements, it's very clear that big companies, Big companies have sort of moved into some other game.
I hope there's a lot of things that can be done at least at Apple's.
Unfortunately, the news is more expensive than real change and in time, some news is out, especially if it was a little earlier than your direct competitor, she would either, uh, ruin the painting of a competitor or, You'll be able to earn some points, respectively. I'm actually a little sad to see all this. It's clear that no company needs to, uh, so, let's say, give some anthropomorphic stuff, like, there, honesty, that's all. Ah, there's one time, remember, there was a story that Google is a good Corpo-corporation. It's actually something like that. I still have a ski on the OpenAI, but I think I'm gonna be hanging out there, though I think you can just stick it out. But what they do is do their presentation for, there, just the day before the launch of direct competitors, and in fact, they don't roll anything, but they just show that we're doing it, we can do it. It's a little bit uncool, in my view. Although again, I repeat, as purely business, there is no question in this regard, but I still want to put on all anthropomorphic skis like that. Here.
I've been watching an interview with one of the partners Y Combinator's management, and they're discussing it, too. What, uh, uh, uh, well, like, the start-ups are trying to play this game now, too. That you don't tell me about the results, but that's what you're gonna do, but you know, make some splash announcement. And, uh, that's really what start-ups don't have to do, because, uh, this is a game, like, a PR business of big companies. The start-ups should, for example, try to focus on the value and, as soon as possible, get the insites from their users.
It's a global trend in the struggle for attention. We'll get some attention, and we'll figure it out, we'll show you something. But people like everybody else, they're focusing on a company like customers.
I agree, yes. I mean, it feels like, uh, big companies, they decided they'd focus on not revenue, not retention, not on the user experience, but on brand-marketing and on winning mind shares. users. Normally, I think it's never been working in the industry, and we're probably going to do the same granites that happen in every bubble.
In general, the same way, yes. You remember we were recently discussing the manipulation of AI-Apocalypse. It's the same thing, yeah, when the company specially blows up some infowater for, uh, getting attention to itself, making some decisions, and also just trying to make such public glasses and sticking around-- You can get your share of the market, not the market, but the attention, yes, which is. So, the fight goes beyond the market share, like, end-users, but the attention that is in
the media.
Well, I guess a lot of people are getting that, so if there's a lot of companies going to go to the Openai now, too, uh, uh, uh, w--- probably what's going on, they don't really understand their strategy for a lot of time. The end doesn't understand the way they're going. Uh, even if you take the OpenAI, they first released one version of the flames. Uh, these plaguins are completely unworking. They released a new version of the plagins and made this kind of corporate workspace, yes, to work together, uh, with these decisions. It's just some ideas. And I think I'm on my own, and Elnar and I, yes, together, and we have this workspace, and we've been doing some kind of plastic there. But the point is, it's more like a very, uh, raw initial story, which, well, it doesn't really have any meaning yet. Although I have a few, like, plastics, and I've been doing some recurrent requests. I'm so tired of hallucinating and, uh, frequent changes in the system of work, there, ChatGPT-4o, that the Yaya, uh, started making and keeping them all up with a bunch of text descriptions to... I'm sorry to use them. I don't know if I have to do something on YouTube, there's subtitles, I have to deal with flowers, something else. I mean, I'm an inter-- I'm doing this plastic in five to seven minutes. It's a totally easy story, right? And, well, if you know how. Iaa's gonna use it further. But it's not a business strategy. I mean, they can't see it coming. So they're the first plaguins they've done, they've thrown out, the second one is now. I think the second plaguins will be thrown out, too. Then they give periodicals these vertical news from the series: here we have done, so education integration. Then it was time, there, they released the tariff, there, for the university. And we bought a company that specializes in c-vide conferences, uh, there, uh, Multi, yeah, it, or what it's called, huh? Then we bought a company that's used to search Rockset. That's kind of a...
independent thing. Then they're issuing news about-- they're making, uh, PricewaterhouseCoopers, uh, contract, uh, what they can sell ChatGPT for 100,000 people there. Something very-- well, if you see business, it's like, well, it's like, uh, it's a little something, like, "I mean, it's usually happening, "something little things that people without experience, you know, "Oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh, oh PR services that are clearly in OpenAI PR are very serious about spilling the outside that they can't spill. I think it's true that this company should be the only one with that investment, right? Because it's too much, yes, the attention of the world. It's power, it's a resource, it's all countries watching it. I'm just saying, "Oh, yeah, yeah, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry. - Oh, my God. I mean, I can't see the strategy. So on the one hand, we know the OpenAI strategy that they are creating AGI, and they want to create AGI, and they want to do so. Oh, okay, no questions. They're probably going there, and some peddata and c-- even fouro peddies are just, like, ChatGPT 4 just, you know, four, he's, uh, obviously, there, faster, cooler, a bunch of things doing.
It's faster, cooler, there's a lot of things going really fast. There's pictures, files, no questions. But these little strategic stories, they look like they're very strange. And I'm actually doing this because Apple, Apple and e-e-e-e-e-e-e-e-e-e-e-e-e-e-e-e-e-e-e-e-e-e-e-e-e-e-e-e-e-e-, that's a very similar thing. Here.
Maybe it's like things change, but it smells like that.
100% agree with what I said. I mean, before discussing the OpenAI issue, I wanted to point out some positive aspects, though. They still have more than 3 billion revenue. Clearly they can sell their models to enterprise. They were also able to spurify, uh-a, the speed of the model. But I don't know what the future of the company is. While the future of the company looks like they're still training-- they're starting to practice the GPT-5 model, which should have a intelligence level, PhD, whatever it means. And it's actually exactly the same level of narrativism that Anthropic's ear, and the same level of narration that was a year ago. So it doesn't change from this point of view, and this narration does not answer the questions asked by the public. The public was asked how they would convince them that the model had no hallucinations, how they would convince that the model was actually studying causal relationships. I mean, basically, if we have, uh, Donald Trump and Ivanc Trump, yes, that's what, uh-a-a-a, Donald is Ivanki's father, Ivanka is Donald's daughter, yes, for example, that the model really learns these. the relationship, because she's often confused, she can only guess one thing, and so on. I don't think there's any answer to all these questions. And even them, they're one of their top, uh, scientist gave... After Soozkever left, Jan left, they-- one of their top scientist gave an interview, and it shows that they didn't do a rehearsal about how to make the model better rhythm. I mean, it feels like they're stuck, uh-oh, like the same paradigm they had in the previous year. And I feel it, including Sam Altman's interview. Like a last year, he was interested in listening. There were new ideas, some ideas were crazy about the chip factory, which was $7 trillion. But it was new, fresh ideas, and they would be clear-- and it was clear what direction they were looking at. Nobody's ever said that before.
I haven't seen interesting new ideas in the last six months from OpenAI and Anthropic.
Look, well, what you're talking about the progress in the OpenAI, you certainly see from the market as a professional in the market, in terms of $3 billion revenue, right? They have a lot of progress for ordinary users, too. And as I said, there's four, it's clearly more and more. The question, uh, is, uh, that, uh, that, uh, that, uh, you-- what happens if they don't move on AGI? Or it won't work. Look, five-- ChatGPT-5 is gonna be AGI. Well, at least, like a discussion. And now I, here you are, Sasha, speaking today. It was yesterday with one man. He said, " Well, they're gonna release ChatGPT-5 now. " And everyone has a feeling that ChatGPT-5 is just gonna be some peddith. I mean, there's no space, uh, space. Well, at least people are not the space, uh, that we've been discussing before. And again, these are not the projects that in the OpenAI in terms of, uh, there, $7 trillion to start a chip factory. Maybe it's going to turn again, or something, or it's gonna be a revolution. But if there's no revolutionary change, theo, uh, interesting how they, uh, are gonna be, uh, er, and how the ChatGPT will be modernized. I think there's a lot of opportunities in there, right? I've said many times that there's a way to turn some unbelievable cool infrastructure inside ChatGPT. I don't know why they don't do it. Uh, in terms of-- there's no question of risphosus. They could have turned this user story in parallel. What's the difference? It's a big business. Very serious, huh? And they could have been doing serious technological things, there, reservoir, there, a major technological thing, up to, like, factory construction, but still doing ChatGPT, uh, cool, uh, cool, uh, cool, uh, cool, cool, ChatGPT and people's questions ask, and memory works really hard, and, like, a lot of vertical applications to make simple users all feel better. Because what you see? That the ordinary people in ChatGPT don't get easier to work, uh, not getting any sense, that, uh, well, I even have to do something, there, or many of you have to do something... Like Sasha said last time in the series: "I think I'm a common user, who writes video proms, right? Although it is clear that the level of proms for the Sashi video is running away from a million people on Earth. Here. And I wonder when people start doing these apdatas. It's like Meta released, so the pdddith, released the pdddette in WhatsApp, there was a I-Posske and everything. That's it. What's next?
Well, there's a, uh, search. Amazon, by the way. That's a great subject, by the way. We don't have it for a while, but I'm s-- but I think it's showing again that all companies are coming, everyone's gonna have it. Amazon has a week, yes, back. How much, Tanya, I think? Maybe two weeks.
Somewhere, yeah.
Two weeks ago. No sharp announcements on TechCrunch, 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, 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, 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, no, no, no, no, no, no, no, no, no, no, no, no
Yeah, I'm just inside--
Inside Amazon. No bet, nowhere. They changed the menu that Amazon is not characteristic, right? It's important to understand that Amazon is the second searcher in the US after Google. Well, it's-- even though it's e-commerce, it's an Internet store, right? And they're getting down there, so they've expanded the number of buttons, which isn't my-- I'm kind of uncomfortable, because, well, you're used to buying a lot, right, Tanya? It's not comfortable. Why is that even? I mean, they did it right, so the prest-- that's an interesting thing. Usually, I know that in Instagram, it's some sort of head of Instagram, there, creator, he's told a lot to change, there's a button, there, I don't know, they can lead to one thing for six months. Change the button. And then Amazon poured out a new AI button, and he's actually doing something. Well, I've been messing around and I'm out of it. I don't know, Tanya, you're using it?
I didn't even get it.
Oh, you didn't even get a ball? Well, I'm like, this one, like...
Not even.
...a wild man...
Like a researcher.
Like a researcher, yeah. Like the researcher got spoiled, but I asked her some things. I realized she doesn't know, she can't analyze my old purchases. So she can recommend to me, this AI can recommend new things to me, but he was alive enough, that's, with some, there, nuances, yeah, but he was pretty alive. Just can't do the data you had. But he can, because the first thing I did, I asked him to order, uh, and take all the orders, like the oil I ordered, and there, olive or gui, there, there, is something, right? And he said, "Find me everything I ordered and compare it to me." Well, I had that kind of request for some reason. He said, "I can't-- I can't get in on your orders," right? And once he can get into the order, it's probably gonna be an incredible breakthrough in e-commerce.
And it's clear that ChatGPT can't use that, because he doesn't have those, you know, he's still not getting that, so much data.
Look, this is an interesting thing. I'm driving in the car, saying something ChatGPT, which means I'm talking nonsense. I'm telling him, "What the hell? Say it properly, professionally. Recheck the data. You're telling me nothing." Half of me turns around and says to me like, "Sasha, can you talk like that?" And I know what it feels like I'm a staff member, so I told my own. I, I said, "Paulina, it's a computer. I'm just giving him an emotional phrase and so he can understand through some emotion of my words that we need to change context." And she had a feeling I had no right to do that. And at this moment...
Does he recognize your emotions?
No, I used some words to recognize my emotions because until ChatGPT has removed the recognition function, there, voices, emotions. But they say they'll be on that side. But the point, the most important thing is that it was Pauline's reaction. She reacted like she was trying to protect him. I mean, usually, well, how do people react emotionally, from the series: why are you wearing it? Well, it's usually a-- that's four kids, that's a lot, it's moms, often it's common. Why did you... I turned around and said, "Paline, it's just a computer." It was obvious that she had this psychophysical emotion, yes, on this subject. And that was really interesting, wasn't it, Ildar?
A year ago, there was a memorial on this subject, when she said there was a separate category of people who always have a healthy life.
Who always greets with chat-bots and then says to him, "Thank you, please." And it's like doing it in case the cars get up. And you've been polite with AI, and that's what makes your chances of survival in this case.
Look, I say thank you periodically, but it's just a machine, I don't think I'm-- I'm not interested
in that theory, am I?
It's me...
She'll remember, she'll remember anyone who said thank you, right?
If we're gonna go back to, uh, the subject we started with, right? It feels like companies are still looking for skikes. The same OpenAI, buying different companies, buying different starts, looking, and where it would be convenient to build the models that she has to make it comfortable in the market. Use it, right? Here they bought a company called Multi or Multi, which is right there, uh, videoconferences, um, specializing, right? And there might be a condom with a built-in, uh, a built-in model that some new espionage would give users. Because it doesn't look like a separate, there, an annex ChatGPT or a separate ChatGPT site, maybe there's a long time of user pick, right? So Copilot+PC is emerging, where OpenAI is built. Therefore, Siri is somehow interacting with OpenAI and watching, and where could these models be built to make skies and users comfortable and to use it? Don't come in or bring your data inside, uh, ChatGPT.
Look, this is weird. They're either afraid that such operators are all like, uh, well, it's Zoom or Google Meet, that everyone will use their models and not connect ChatGPT. On the one hand, they're afraid because a little strange story, too. Well, what are you gonna be doing with these companies of a different type and build them? Or are they planning on doing this just like that? I think again, I have a very simple parallel view. You know him. I say that ChatGPT will, in the idea, be new operating systems and within these new, you know, ChatGPT can be his own operating system. Suddenly, it's not ChatGPT, is it? OpenAI may have its own operating system, where a man will work and spend a great deal of time there, because in fact, he will not need any other interfaces. So, in the chat, on the basis of this, yes, you can turn a lot. Maybe they're gonna do some tests for that, you know.
Oh, oh, oh, that's very interesting. You have a point of view that is completely different from mine.
Although I am not a legitimate man, I wanted to present my point of view. Mm-hmm, I just think the OpenAI has ended, well, as the speed of innovation with LLM models, it slowed down a little. And, and not-- I'm not saying we're not going to reach AGI. I'm just saying they want to rock models further. And the next generation of models takes more time, requires big clusters that no one built, and so on, right? And that's why I don't know. And I think they've been doing what they've been doing for the last six months? They're trying to figure out how to become a business, how to deal with their own new nun-profitis story. I think I really want to believe that they're doing it and trying to figure out, uh, whether they can be a public company, and so on. Ia investors don't like consumers, and they have half of the revenue from ChatGPT, with the Consyumeric direction. And I think what they're doing now, they're doing in the first place trying to buy more, trying to build more like a software function. And both of these companies, Rockset and Multi, it's like a software function that's needed for enterprise, you know, for business decisions. And I think for the OpenAI they want to build more affordable universal business decisions. And besides the castom enterprise offering what they have now. You shouldn't have connected the outside RAG, right? I mean, well, there's a business here, so they have, they want to use the OpenAI. Not that they should not know how RAG is going to connect some outside, pick it up. It's complicated. There are three companies around here that are hard to understand what they are. And that a decision from OpenAI be made.
So this is Rockset.
I also support because, uh, it could be a consequence, yes, of what they understand is that AGI is more than it seemed, and the current level of theirs just wants to stumble on other decisions that will be made Use, right? Just assuming there's a year, a half, two, maybe, and there's no need for a new Zoom. And if there's a feeling that rocking or whatever approaches are in place is something new that can't be done, then there may be a point in the current market to take a part of the market into, not just, Reserve, huh?
So they're taking him. So they're taking him back, selling it. They-- they have three billion revenue, and they sell a large number of B2B companies, uh, their system. And when you buy, uh, other, other... Let's just see what happens to this Multi or Multi, yeah, Multi, what happens to him. Uh, is that even gonna exist or not? So they're buying her corn-- to just swing on the outside market like a videoconferencing company or whatever--
And if we're inside, then we'll see, uh, implicate, they're just buying to use it more. We'll see the appearance of a huge number inside... Well, again, that's my subject. Well, in terms of my view in terms of the operating system, there are a bunch of applications, their structure, well, their infrastructure, a huge, huge infrastructure of applications. And why, on a parole basis, I'm gonna have to go to another system, some kind of system, right? And, like, if you take Zoom, I always said there was a moment, I remember, Zoom was worth $100 billion, right? It's, if I'm not mistaken, September, autumn, stew--e-e-e-e-e-e-e-snow, autumn 2021. And I'm like I-- it's always been a very strange story, because Zoom is not the same sé-- that sopht that makes the user, uh-a, stick to himself for life. Well, that's a ChatGPT, right?
I can. ChatGPT can be tied to the-- for life. Zoom you can't tie your life to life.
Yeah. I mean, I think the next steps are logical for OpenAI. This is, uh, Team Plan. I mean, we're gonna count Team Plan as a SMB type, add, uh, RAGs, more collation, then there's a personal assistant, and so on. And I think, you know, and they're gonna be, I think, the target is like-- I agree with Linar to raise, uh, my presence on the already existing marketplace of chi, you know, how to pick up the whole, uh, B-B2B thing. To my own. And there's a little more to back off. I mean, I think, after all, that AGI people, I think, in fact, AGI needs to be understood, it's when AI can improve AI. Here. I mean, it's like AI agents, that's, like, models that improve like a front model, right? And that is what AGI can say because it's more like human influence to improve models, and AI can work with AI. I'm doing it, AI can... run a lot of experiments that people can't. I'd call it AGI. But! Before we move to AGI, there's a point where some other type of work besides customer support is automated. Customer support, uh, I think Intercom, uh, did it more than a year ago when they said we'd be charding that we're introducing a new final model of coinization, that we're not looking at a subscription, but we're gonna be a little more excited. I'm sorry, I'm sorry. I think the next milestone is agents. I mean, agents are when you hire a staff member, but you can't pay for the result. So almost nobody works. Everybody wants a paycheck. And AI will say you pay us for the results. And then the next step is when more work, not just customer support, is going to be on agents. In customer support, statisticsa suggests that the number of people working in customer support can be reduced by 60-70% by the generative AI.
And when we see this traction for the lu--- for other types of work, then you can say that we've reached the next-
You think there's gonna be a payment? I'm just having a hard time telling you now, to imagine that there's a SLA in the customer support, right? Service level agreement, that is, some rules that the staff member should hold in terms of client relationship. In time, answer, in that place, every customer satisfaction, so, well, uh, get some feedback from him, there, deal with his requests, tasks for a certain amount of time. Oh, I'm honestly speaking, in business, I think that's, uh, the moment when you can take the money for the result, he's migrating very fast just to the soft and the operation. Again, maybe I'm wrong because to take money for a pure result, someone has to have a unique agent technology. Either OpenAI and all such companies should prohibit all external companies from creating agents, they will create them themselves and sell them for results. Or in all companies, it's just gonna be a use--
Or all companies will just use the sophth. Well, as such, it's just a function that's gonna be writing that's there right now in the same customer support. I don't see the case where the agents, you know, how an independent unit is sold. Because if you put, uh, a conditional check--- uh-e, the company pays for the salesforce license to one sales manager at $50-100. If they come, they say, "Pay, then pay more money to improve the operation. But, well, how would we like to reduce the number of staff?" I don't think we'll cut it. I don't understand how the young economy starts to fold up. But the subject is interesting, by the way, and it can be raised separately. It's actually big, very interesting in terms of those agents and discussions.
Maybe I'm wrong, Sasha, I heard you from the model, but you said an interesting model. It's a fun subject.
It could be a little wishful thinking from me, right? Because I'm gonna explain where I wishful thinking comes from, right? So globally, we have an inflation-- almost, I think inflation is more in every country than GDP growth, more or less, right? I mean, we're actually in the recession, deep, there's been two years, there's more than two years, and we're-- we're-- we're-- almost the world. And, uh, so I want to believe that generative AI can make it a new level, like an economy. But I think that if it happens, it should be an innovation not only at the level of technology, but at the level of business results. And I want to believe that, uh, AI will indeed be able to improve economic efficiency. And there'll be skies where people can pay for the results. Because, after all, enterprise-sale is, like, a boloto. I think that American boloto is when, uh, uh, well, actually, there's a lot of sophth-soft inside the company that they don't use, they don't know how to use. When the salesforce is made very difficult to build, the integrators are hired to build it. I think it's a very difficult one, I think. I think it's very, very complicated and it's not effective.
And how to sell the result is how successful generative AI, new generative AI enterprise-stage can replace as incumbents, yes, how they can compete. That's my point. I can be wrong.
I agree with you, by the way, because I was working at the company where we used Salesforce, and I remember that was a bit of a frustration. I'm honest, like all the other employees, I didn't know why this is all about writing every step of my own, some weird stuff there... Well, I mean, if there was some AI that did it for me, it would, well, make the whole procedure much easier. I mean, there's actually some resistance on the part of users to use this program.
And the same thing with Atlassian. Any big enterprise company. They sell one product, but then you have a lot of food. You can't pick one, you're gonna have to use it all. And there's a lot of chakovs like that. It's like social media. A lot of hak, like to make a concuemer that a user enterprise, no matter how, not to leave a decision and, uh-a, and, and actually, to sell more of the jobs that companies often do, and, uh, 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, 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, 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, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, And I think, uh, that, uh, that, uh, in these, uh, gray growth tactics, if that's what you call it, yeah, and that's a huge value of construction, there, really big business business business ventures. Now. I want to believe that the generative AI will be more about real efficiency in the end, not about these hire methods, there, sale and how the hacking decision to make an enterprise. But for now, of course, there's not enough for this signal than customer support.
It's interesting to have to go back to the news from Pricewaterhouse, right? Aah, the news was that not only was it that there was an enterprise plan for 100,000 employees, PwC got from OpenAI, but also the possibility of selling decisions on the basis of ChatGPT agents to its clients. And I wonder if PwC can find some, um, skirts that really will be useful to them, users, yeah, their customers. Well, I remember, they're auditing, respectively, data analysis, some other stuff. And, uh, maybe the company's already been hit or it's still thought that there's something interesting to find on it, uh, earning some decisions, some agents, again.
Yes, or there, selling services.
They're selling. Look, all, yeah, all the big consulting companies, they just cut the money they come into a big company. A big company is sitting on the board of directors, saying, "We need to introduce AI. So where are we going to put it? Come on, let's ask, let's ask, so big consulting players." They start paying millions of dollars for that, for consulting about what they're talking about. And that's why this topic, I think, is an increase in revenue in B2B. Well, maybe I'm just like this...
So, in fact, PwC's clients were too often asked, "Let us some AI," and, in fact, they found this way, right?
Yeah, 100%. So the consulates are the most likely to run from it now because they already know how to sell. And no, well, no company that gives a complete enterprise solution. Not even OpenAI is his RAG, is it? So what's the talk about? And of course, PwC is really strong. Bene-bene is huge beneficiaries, all the consulters are huge beneficiaries of the AI-Volna.
Look, I think today's a meeting and a really interesting, a little different. I'll remind everyone who's watching us that our channel is called ToTheMoon. We go out once a week, we go every Sunday. Uh, we've had 13 episodes. At least today is the 13th edition of the day we record this video. This video will be coming out too. Hold the canal, click, sign up. That's all it allows, uh, to significantly improve our dynamics in YouTube. YouTube needs it. Keep your comments, too, share your comments, um, your opinions, discussions. And those topics that are interesting are interesting to discuss in the future. See you again! Thank you very much, guys. I think it's a good, good morning for you and me. Bye, everybody!