But I think even the size of such investments, it's gonna fall down there, from 100 to 200 million, as it was before, maybe three to three to three. You're doing something on the entrance, then some magic inside is happening, and you get something out.
The number of work for programmers is only increasing.
There are always two good rules on which start-up can be launched, right? It's like they're making antireculum themselves.
Who's gonna give a billion under this thing? Why would they give a billion?
I think that in June of the 24th year, the investment market in generative AI was changed. It's hard to get investment just under, uh, model development. There is still, of course, an interest in investing in the foundational model, especially when DeepMinds do so, and when they can refer to the resurrection they have already made in the industry. But I think even the size of such investments, it's gonna fall down there, from 100 to 200 million, as it was before, maybe three to three to three. I think investors are going to look for companies that, uh-a, can get 50 million and 100 million revenue in generative AI. And these companies, they'll be receiving still rabid multipliers, there's 50 tyre 100 to revenue. Uh, but now, I think, the VC's main question is, who's the company that's gonna get to 50 million revenue? I think that-- everyone saw that the company, like HeyGen, had problems getting rocked after 20 million revenue. I mean, uh, a lot of ears in the industry seem to have reached 10 in a tyre 20 million revenue, but for different reasons, revenue is not super high quality, it's more like innovative budgets. Im-and-and-and-and-and-a big award will be paid to companies, I think they'll reach the 50-mile mark of 100 million revenue. And those who can prove that this is a high-quality revenue.
Look, if you're looking at a little more about a normal start-- about ordinary people, not those who already generate, there's 20 million revenue, right?
Aah, who starts the launchers. So how much chance do they get on the account that they're starting, that someone's gonna believe in them now? I understand that the start-ups actually got bigger in AI, and more competition, more complicated. If you've ever released some idea, it was easy to even put it in a corporation, now that you're being approached and offered 15 service in the safe, there, 20th service in marketing, you're not that big and you're not. You'll test it. How easy is it, you think, uh, you know, to start a start-up from scratch?
I think there are always two good rules on which to launch a start-up, right? First, when there's a successful product in the corporation, and for him, the para--- it's been possible to make the whole tooling. I think it always works well, to go and make that tooling accessible to all. Yeah? I mean, it's like the first standard way to make a successful start-up. And the second standard path is when there's a deep understanding of the problem and there's a deep--- and there's already an understanding of who the first customers are, right? Uh-oh, but I think it's hard to do a start-up otherwise.
You can bring it, you said the first way you're gonna use the tooling, just a
few examples, so the people who look at us understand, uh, get it, uh, uh, uh, you know, uh,
Yeah, yeah, sure. Uh, I mean, there's a-- a simple-- a good example-- it's Temporal. I mean, of what I understand T-Temporal is a-- it's a platform that allows microservices to be orchestrated. I mean, it's a infrastructure startup. And I understand that, uh, faunder, uh, Temporal, he used to work at Amazon before, and that was why I think it was very, very, very, very deep, understanding of how much trouble would be, right? And he saw that there was a possibility of creating a separate independent c-company. I mean, when you did-- for example, if there was a company that would do well, inside itself, would use a lot of LLM, right? And then, the engineer, the engineer, the engineer from this company, will go to make a company that allows us to do what? The Rowling between LLM, right? That might be a good company, because the faunder had experience with how this rwling could work. I mean, I'm very sceptical about the development of tooling and infrastructure when the faunder doesn't have e-spionage, how that tooling was used to build a successful product. After all, we need a semi--- build a successful product, a tooling for him, and then we'll start a start-up so that everyone can use that tooling. And tooling is a toolbox? I understand correctly? Or how do you know that, tooling? Yeah, tooling is a toolbox, yeah.
I mean, now, for example, there's, uh, uh, uh, uh, uh, uh, tooling, like, nowadays, there's a lot of services to create presentations, that's, uh, that's the same CapCut and so on, right? But, uh, what I'm trying to say is that it's hard to go and make CapCut like a vacuum. First, it's a roll. First, we need to make all these tools to edit videos, roughly, inside, big corporation, realize that people need to know that they don't need anyone to pay for what castomers are willing to pay for. I'll pay you a lot. And after that, to be an independent company with this tooling, but it's for all the clients. I mean, that's what it's like for tooling. Aaah, well, you can always do a start-up that's gonna solve a particular business problem. It's just, uh, it's, it's-- it's always a slave--- it's always a work topic, but it's just that there's no such big award. I think that-- and the point is that there's already been a lot of acquisitions, right? I mean, it's already been seen that the market, uh, start-ups that LLM does, it's already been, like, two years, even more. And it's obvious Stability tried to sell.
Inflection, as if acqui- acquihire made a deal, sold Microsoft. Potoom, Adept wants to sell now. Perplexity, they've already been made by offers to buy them. Humane wants to sell-- they're gonna... they're gonna find a way to sell for a million dollars. A billion dollars, sorry. So Humane is looking for a million dollars worth of sale. So the company is somehow incorporated in the 17th year. Uh, and now, like a peak, that's what I'm trying to say, companies that started in the generative AI, they've already caught their pick, right? And the signal is that everyone wants to sell on this pick that's now. I think LLM has a pick. Pic is like an excitement, a hat, when it's just fun to do something. 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-th---a-this, that' that' that' that' that' that's-th-th-th-th-th-th-th-this is-th-th-th-th-th-th-th-th-this is gonna be a video understanding. And I think there's gonna be more acquisitions in a video understanding right now, and in a year, it's gonna be like a peak of a hat around the video generator. Sora's coming out, Google-model's gonna come out, like a peak of the Haip around the video generator. And then again, only two or three companies, like, will remain in every field, and the others will either close or beer or sell.
You said about Humane, uh, million dollars. I thought...
Milliard, billions
...so little at the beginning. And then I heard a billion and I thought, "Who's gonna give a billion in that thing? Why would they give a billion? " Just a interesting story when they released, uh, video, I think it was a year ago, and that they're gonna be launched before their, uh, first-- for their sale. It looked fun. And then, when you worked and talked together, there, chat-bot, LLM chat-bot, you realize at some point that you don't need a device like that.
So I wonder who they'll sell at all and how much.
It's very, very difficult to imagine, yes. I mean, it's hard to imagine, too. I mean, in my understanding, of course, acquihire deals, they've always been. If the standard size of the acquihire deal was in the area of 100-150 million, the too is, with the inflation, it could certainly grow to 300 million, buto is hardly more. I mean, I think there's a business law. I mean, yeah, there's like these technological waves, and it's hard to break it, right? I mean, it's very, very difficult to... very, very hard to break it. I mean, yeah, mean Adept AI, a company that started to do multimodal models, they're now trying to sell 500 million tyre billion. And I'm pretty sure they'll sell, uh, until the end of this year. And, uh, again, this wave that starts buying, uh, start-ups that do multimodile, uh, because for a lot of, uh, those who do video analysis, advertising and so on, uh, decisions for video understanding will be important. It's a pretty complicated forensics, and the Adept type of start-ups are definitely gonna find their home.
Good start to The Moon. It's a good subject. It's interesting to be, uh, in the middle of something like this and to see how it happens, uh, and how the market changes, how different currents go. Ah, Ilnar, look, you just told a super interesting thing--
What was the super interesting thing that we used to do before we turned the tape on, uh, the research on Anthropic. Uh, you can-- and these examples of changes, uh, neurals inside the neuronica about her understanding of the Golden Gate Bridge in San Francisco. Can you tell me a little bit about this research from, uh, Anthropic?
Yeah, sure. I'm really excited about the wild. I read it straight with a stubbornness.
Here, er, gentlemen, read who's gonna have a wish, um, Ah, 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, 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, Let's try to say a little. Look, uh, when we work with neural networks, like, with GPT the same, yeah, with Claude, there's someone else, uh, there's this black box concept. You're doing something on the entrance, then some magic inside is happening, and you get something out. Ia, you don't really understand, but how did you do, yes, that you sent him some kind of request, he gave you some kind of answer. That's more, uh, not only do you not understand, nobody understands why it works like that, right? I mean, the internal machinery itself, how the system changes, and how it is, it gives, no one understands it. And that's a problem, yes, because you can't control the whole thing. It's a system that's so unused, as it says. I'm not sure how math is a big deal to understand, because math is all evidence, they're like that, right? Here, Pythagoras once proved the theory, and now any man can check it out and make sure it's true. Aaaa, with neural networks, it doesn't work, does it? Two people can do the same prompt, but there, with a little different system conditions and, uh, on, on, on the way out, there might be different results. So, accordingly, this system in the form of such a black box, it's a little uncontrollable. And now Anthropic just tried to shed light on what their own neuron network works. They're talking about a pretty interesting method they've done. But, uh, during reading, there's always such anthropomorphic little associations in their head, yeah, like a neural is a brain brain. They're finding different areas of this brain, and they've learned to find different places of interest, some combinations of neurons that respond to different, a-a-a-a-a-a-like signs. Well, like when it comes to the Golden Gate Bridge, which in San Francisco has, they have one zone. When they say something else, some other zone is being lit. And they found the pattern. It turns out that there are other concepts near the bridge, which are somehow close to the bridge. Yeah, there's an example of Alfred Hitchcock's film that was filmed in San Francisco, and there he is, too.
There's something else related to bridges, too. I mean, there's some topology that you can find. But it's like one part. The other part is that area can be changed a little, yes. If you have access to weights, you have access to neurons, you can play a little game with these neurons, yes, you can vaporize this brain that works. Here they tried, uh-oh, this area is a little tightened, yes, to increase these weights. And that led to quite unexpected, uh-oh, consequences. Neuronic who used to communicate like normal, started being a little cycling on this bridge. And when she's in good shape, there, is, "Claude, what--what's your form?" she, uh, was standard in there saying, "I don't have uniforms, I'm AI," and so on. And when the zone was lighted and slightly increased, she started saying that it was in shape like the Golden Gate Bridge, and, in general, it became cycling. Well, that's a little, uh, weird, what's that for? Uh, there are other examples they're giving up there. For example, they found a zone that was responsible, to say, for lying. And in general, when the system is trained, it's being trained that some subjects can't be answered, uh, that's not true anymore. But if this area is slightly stiff, it can be made, for example, by writing, uh, by this N-N-- using this neural network or bypassing any of the restrictions that have been imposed. Yeah, I mean, basically, if you have access to the weights and you know the topology, uh, this neural network, you can influence how it works. That's the first. And the second one, you can create these, I don't know, mutants, some kind of thing, right? There, genetically modified neuronets can appear, which, consequently, may be reinforced in some direction and vice versa, they may have some gene, yes? But it's like, you know, how homemade animals exist. There are wild beasts, yes, the gene of aggression is in some form. And there are domestic animals that, in the form of a sedition or something, have them have this gene of aggression, it's wrapped up, right? And so, there are cats, home dogs, they don't throw at people with proper education. And whatever you do with wild animals, they can still have this gene and, there, aggression can grow. And some of these, um, associations, some of them, uh, uh, uh, uh, they're gonna be associations, they didn't-- they didn't leave me all of my reading. I therefore strongly recommend that: read, it is a very exciting publication.
A
cool story, of course.
It's like they're making antireculum themselves.
Why?
Well, because, like we've created something, uh, and it's kind of under some law and order. But if anyone gets access, uh, and you can influence this system, uh, then the result is bad, there, and it's not known what's going to get out. She might start lying there, talking roughly, "I dare to act aggression and so on. I mean, well, that, um, I don't know a little. Elnar, what are you gonna say about this? No?
No, no, they're on the article themselves saying that there's no need to be scared. It's clear that you can change weights like this, but on the other hand, it's easier to just retrain the model, right? What if you have access to the model's weights, even not necessarily knowing the topology, you can just get it to some examples and, in fact, remove these limitations on the lie, something else. It's gonna be a lot cheaper. It's more like a, uh, scientific study to finally understand, but this black box works, yeah, like this black box they created and which somehow fits right into some kind of question. Why is she doing that, huh? It's like inside. It's more like a, um, some kind of scientific research that might be spilled into something, maybe not. But it's still a little bit more pleasant to know that people are beginning to understand, and what they've created, right? Why does he answer rightly and why she's here making such a z--e-e-e-e-e-e-e-e-e-e-e-e-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s
At the same time, they seem to be giving her some human qualities.
I was just saying that these anthropomorphic verbs, they're bound to be taken into the skirts because it's still a set of functions that produce with numbers. Iia with Golden Gate bridge is exactly the same. I have to understand that there's just some weight, they're all raised, and that's why the answers just-- these numbers are bigger, right? And that's why she started to stitch her. T-that's also a thing to understand that these anthropomorphic comparisons with animals or something, or, with a man, they're all obviously in the skirts. But the study is very interesting, in my view, and important.
The most important thing is that they're creating something. The most important thing is that they're building something at the beginning, and then, uh, they say it's perfectly normal that we're doing the research presentation here. That's what this is, uh, that's what it's asking, what else is there, where the research didn't?
It's clear that there's no access to, uh, weights, same Claude or, uh, OpenAI new GPT, or any other big neural woo people, right? And there, for example, er, some villain, without access to them, can't do these specific operations. But on the other hand, there is an open source of the model that can be subjected to the same research. Yes, they are probably expensive and all, but it is possible, therefore, that these studies, these approaches to neuronet research will be further developed and some open models are already in place, I'm sure you can make these changes directly, you know, or just explore how they really work.
Well, the developer has more chances of putting his personal change into the system, uh, that no one's ever gonna find.
And that too. But I still hope that neurons will come from, uh, compressed or, uh, underneath the genome of aggression, for example, or who can't lie or something.
In the future, this AI will be used, and then, uh, whoever gets access to them has more power and capacity to harm, subtle there.
Or a good thing to do.
Well, or like domestic robots, like domestic animals without the gene of aggression.
Let's think positive.
I want to continue this. Uh, Microsoft is a lot, uh, updates of different information made and there's a lot of information about updating them inside, in the Corpo Corporation. Before we go to these updates, that's the question to ask. There was a time when they were so close to the market as if they lost and were not heard, and now they're the most expensive capitalization company. Ia $3.2 trillion dollars. Apple is just below 3 trillion and NVIDIA, and it's 2.6 trillion. So NVIDIA is running Apple soon. It looks really weird. Well, it's just, it's just a little weird, right? When Apple was out there, there were some questions about it. I guess there's some sort of question now, but now NVIDIA-
No questions, but now NVIDIA is 2.6 trillion, and Microsoft is 3,2, at least here for, uh, today. Ah, Microsoft is this growth in stock makes them real innovations or, or do they have a resource that has a NVIDIA? Ah, still, it's about the resource. Why? Why is this a success? We're gonna move in there, to innovation or to updates, but I think the updates they're doing, they're all in the market, they're in the cool, but it's just the cost of some space thing. It's like 35 or 33% of the Euro Union budget. GDP, GDP, not budget, but Euro-Asian GDP. I think the Eurosus GDP is 10 trillion, if I'm not mistaken, or something.
I'm surprised how close NVIDIA is to them because for me, Microsoft is so straight-up hegemon and operating systems, and that's just the dextop versions and server versions, and a lot of them. Apps that exist. And, uh, and, uh, it's a huge market hug, anyway. All that's about the P.O., and all that's got to do with Microsoft. Like a user.
Well, of course, the world's largest operating system. Sure.
Yeah, yeah, yeah. Here. And NVIDIA, which, well, roughly, only has video cards. It's clear that it's all computers, computers and stuff in the hip now, but they're still worth almost as much, right? And I'm a little surprised. I always thought NVIDIA should be a lot cheaper, but it was worth it until all this story started. But here.
Ildar, it's worth a lot less than $600 billion.
It's a big difference. Yeah, yeah, yeah, yeah, yeah, yeah, yeah,
yeah, yeah, yeah.
But the number is there, the number is there. That's interesting, because when it comes to change, it's a change in resources, including. Who's got a resource, yeah, who's got the possibilities. As you say, the more you stand than-- the more you start to cost, the more you start to cost, the more you have the resources. The more you have the resources, the more you're standing, huh? And this endless cycle.
I'm not really a economist. I mean, it's very complicated. I'm trying to compare, uh-oh, now I'm trying to find out how Microsoft's financials compare to the rest, but a few more b-b-b-bears are like, like, superficials. I mean, I think, uh, Amazon has-- Amazon-- Amazon still has a leader in the claud, and, like, Microsoft shows that they're getting them, right? I mean, I think the main bever is like that, that more than 50 per cent of enterprise uses LLM through Azure, which would suggest that they are in the new segment like claud services, AI services claud, they're a leader, Yeah? I think this is really-- I think it's something that's been disconnected with all the others. Uh, from the point of view of, uh, in terms of, for example, enterprise value on sales now shows that Microsoft is highly overrated compared to Amazon and Google. I mean, even for enterprise value to EBITDA, that is, Microsoft has a b-mobile multiplier than Amazon and Google. I think, on the basis of what I think everyone else likes Microsoft because they're making acquisitions, right? So all these companies have the same question as how effective they can reinvest the cash they have. Aah, no one's interested in investing in a company that's roughly $100 billion but doesn't know what to do with it, right? Because why if the company doesn't do anything with the cash, it's easier for the s-sama to decide how to use that kash. And the question is, how can a company reinvest? Microsoft did a number of acquisitions, including Activision, which has forgotten their Xbox business. I think the main problem is Amazon and Google that they didn't do good-- and Apple that they didn't make good acquisitions. And this is the main criticism of Apple that they have the largest balance sheet, that they have a huge cash reservoir, and they don't use it. And that's not interesting for investors, me-man-- less and less interesting. So Apple Stock's multiplier isn't as big as Microsoft.
I have a very simple explanation, I guess, but I'm happy to hear your opinion if-- well, I see it so far.
So, is it that Microsoft is with the appearance, uh-huh, in any case, additional financial opportunities, although I don't know about the financial opportunities, they can get new ones in the future, The possibility, for example, of production in some or, well, of their integration with the OpenAI, does that mean that they will produce more or no more tough decisions for ordinary people?
I think Microsoft has an opportunity, a real way to move now, uh, they have a real opportunity to move, uh, NVIDIA. I mean, I think that, uh, I, I think that, yeah, uh, for Microsoft, it should be a top priority, how to eat all this rabid NVIDIA brand.
Yeah, yeah.
Yes
... moving. In the sense of not letting them go? Don't let them in the cloud or in terms of...
I think I'm gonna go in the chips. I think Microsoft has a unique opportunity to go into the hardware, to go to the chips. So they had a lot of unsuccessful hardware products like Microsoft Surface, they were planning on doing, they were doing the phone, and so on. I don't think it worked very well for Microsoft. They've done good with Xbox, but Xbox is using the outer iron, right? So, there's a c-card in there that's not a microsoft-ovie, and Microsoft can now be under, like, LLM has special chips and a share of the ow--- market share of NVIDIA.
I think it's now they've got such a big opportunity.
I wonder how different-- like, company-to-company, but different companies are basically, right? Amazon, which, uh, second-- main e-commerce in the States and the second such search in the States where people sit, the goods buy, the physics are normal, right? Here, c, there, Google is like a searcher, all alone. I mean, the other kind of environment is for physics. Ah, it's all inside the proo, this huge, cloud infrastructure and the desire to enter, including the production of chips. Microsoft reported that they had released new computers on ARM Architecture. What is this? What's that step? Where's the movement going?
I think again, I'm just gonna say just like a hypothesis, yeah, that they're just using Qualcomm chips and want to experiment with other hardware types, yeah, that's what they're trying to find. alternative solutions, because as long as Qualcomm has historically been successful, and in the VR decisions, Qualcomm is used, Qualcomm andia is used in many Androids, well, I'm not sure what--- how it is. - I'm not gonna say that. Uh, well, Qualcomm, that's $200 billion, right? So, as if the , well, rudely speaking, if-- if Qualcomm proved to be a sufficient competitor for NVIDIA, Microsoft's condition could easily buy it. I think it's important for Microsoft-- well, because it's a big company, they can afford to do whatever they want to do, and explore what's on the market and how it works.
I'm a little worried about Intel. I've had a dream since I was a kid that, uh, Intel's processors are something, uh, desirable, right? I had the first computer on Celeron, but I wanted Intel. Now, it's like, but-- and now it's like a lot of, uh, corporations, uh, Apple, there, with his M-- with his M-chip line, yeah, right now, Microsoft on ARM architecture is moving in and the x86 architecture, which is where most of it is, Uh, Intel chips are built, they're like they're gonna be unnecessary over time, and it's like this time is very close. And, um, first Apple, yeah, they decided that it would be better to move to another architecture. It's like Microsoft now, maybe confirms. And here I am wondering where the world is going in this direction.
Nothing good, unfortunately, yes. And I'm just gonna want to share some of my misunderstandings, right? I mean, it's clear that there's a value not only in the production of chips, but in who uses it, right? I mean, it's generally clear that Qualcomm, Intel and all--AMD, everybody's trying, uh, how to make their videotapes fit for different, uh, tasks that are now covering like NVIDIA, right?
Which surprises me very much that no one and-- that no company is building an app ecosystem around itself. So everyone-- Intel has a venture of ARM. Q-Qualcomm can invest, AMD can invest. So these companies have the opportunity.
The ability to build the seed ecosystems around you. And what they don't do is really amazing. So historically, I'm going back to Snapchat and their deal, the famous Google deal, I think, in the 17th year, when Google Zach made it like-- when Google made a kind of tricky deal like that with Google. Snapchat, that's what the conditions are for five years, isn't it? I mean, they actually found a way to plant them on their ecosystem. And I think it's very surprising that this is not happening in the generative AI. I mean, uh, companies like Intel, AMD, Qualcomm and others, they, in my view, should be very aggressive in trying to develop the ecosystem around themselves because there is a need to literally find one practical application. It could, like, be around the ne-- and around the-- and the super- and the pre--- and defeat NVIDIA. I mean, but it's usually the start-ups, right? Every start-up is trying to find a way to start with something that a big player can win. And the fact that the big company-- that big companies don't make any sense of trying to make NVIDIA feel surprised. It's just, it's very strange, it's very strange.
Is that a management or is that too much money? Or is it people's too--
It's very strange, because, well, it's clear that Intel can allow a few start-ups around himself to incubate, for example, purely, like, this very fast-track, Anderstending video, for example. I'm not sure who's a lot of applications, is it? I mean, I see there's like big models for the Anderstending video, but there's little ones. Intel could be around itself, for example, to build an ecosystem and, uh, to make everyone use Int-Intel video cards. But it's not for any reason. That's very, very surprising. Or it makes sense that everything-- the whole process, there, video processing, code, and so on, that's all kind of a little change from generative AI, and Intel could be under this yuseus, for example, sinking. I mean, how could they do one or two such strong hypotheses and check with five tyranny ten start-ups for each hypothesis?
They have enough money to do that, but they don't do it for a reason.
About the Arm switch as a user, what can I say? Those who have been working for a while, uh, Mac probably know the problem, yeah, that there are appendices that Windows work and the Mac doesn't work under. In particular, this is not only connected to the operating system, but also to the architecture, it can also be linked. So, with Windows, there's a story like new chips that, uh, there's a part of the apps that are written initially under the Arm Architecture, and some of the applications have to be made by emulators, right? Just like you can put Windows emulsion on Mac and, in fact, use some magic-- some Windows-appliance. The same may be necessary for new devices to also be made by Windows users for their own Windows apps. Because the transition to a new architecture is still causing the software manufacturers to do it, too, well, maybe not ready, to release a release or your application at all. as a whole, until, for example, the rewrite is not being rewritten, and so on.
The number of work for programmers is only increasing. We'll have to rewrite everything all the time. Who says the programmers will end, of course, so only see growth, just growth. It's all to be kept in the first place, all of it needs to be rewritten, and at the same time have some trillion versions for all of these systems.
100%, yes. I mean, it seemed like there was an assembler earlier, yeah, people were programming on the assembler, on C, and it seemed like Python was a lot easier to write the code, so less programmers would need to be. No way. It's true that Fortran and Assembler, they may have a lesser scope than it was. Noo's number of programmers has grown. So very-- sure, it's just gonna be new tools and new professionals, but old tools will use fewer people. But in general, the industry will certainly grow.
People who watch YouTube often seem to start, uh, getting a mouse to the tub because there's got to be a promotional integration that we're so careful about and re-smarting it. Just tell you, don't worry, we're not gonna have an online promotion, at least in this video.
That would be very funny. Here. We're on ToTheMoon Channel once a week, and we're talking technology news and sites from the Silicon Valley and around the world. Sign up, write your comments, stay with us. Before we meet!