it's just a zebra, the stripes are a camouflage. the sculptor has his own philosophy, while both creators are united in one thing. even the most different opinions can exist absolutely calmly and not intersect in conditions of complete and absolute confusion, complete chaos. immerse yourself in the paper worlds of latif kozbekov or share the iron irony of andrei bisti. a double exhibition of contemporary artists will be held in the exhibition hall of the creative space new manege.

called kandinsky video, we will continue this line, that’s it, but what makes it better is better knows the domestic cultural code, of course, this is what we actually set for the model what task, and we will certainly develop this further, what is the problem with russian data, here is knowledge of the russian domain, is that there is certainly less russian data in what does the model learn, it learns from data from the internet, that is, from open pairs of images and... writing, it seems

to capture data at the time of the request, no, well , naturally, there is a team that deals with the data, and that is, the data in it already contained, she is not searching on the internet at this moment, no, no, no, in no case, she is creating, this is somehow unique, where they are stored, this voluminous amount of huge amount of data, kandinsky the third studied on one and a half billion pair a picture with text, and you can imagine how many terabytes of data storage are needed there. on a supercomputer there is, as it were, all this is learned on a supercomputer and stored nearby on special storage facilities, they are called s3, but these are such large, large data storage facilities that are really for, say, in order for the dataset - the model to have the opportunity to see, they need to be put there, and not just put there, you need to remove all low-quality pairs, that is, filter and remove everything, well, if you go to the internet, then in general there are half of them. the data will probably

not be of very good quality, it’s all full of rubbish, that’s it. the picture does not correspond to the description that is written there below, or it may be that the picture contains a watermark, yeah, such data is probably also bad to use, because the neural network will reproduce this watermark, she says that artists are already fighting, that is , they are spoiling their image, so that it is not also true, this also needs to be fought, that is, there is a huge number of filters that need to be written before giving the data to the model, because of course you can’t give everything away, it ’s like you know, showing a child everything in a row, it’s not clear what he will learn in the end, of course, the data development team is responsible for this, and this is super important, both the activity and the team in itself, because from her not only half, but more than half of all success depends, and after the data is stored, here they are...

photo editor now, in general, lately it has been very popular to make a photo editor based on or rethink current photo editors such as photoshop, integrating ai there, why is this useful ? because it is clear that creating pictures from text and editing pictures is much, much easier than actually drawing yourself, what people always did before, it took a huge amount of time, is now quite simple write in text, a picture will be generated, you don’t like some area, you blurred it out, added whatever you want, it really simplifies the work of designers, and very much so, that is, the ability to not only generate an image, but also edit it with the same text, yeah, and moreover, draw more than one picture. some designers are afraid of this, that we are sort of automating their work, in fact this is not the case at all, because artificial intelligence is a tool that allows you to simply

optimize work, it applies to any profession design is no exception in general, there is a routine part of this work, it needs to be automated as quickly as possible in my opinion, as much as possible... better, this will only save people’s time, in fact, the quality of the content will increase, and animation, for example, how much is there routine, animation is not about creating a video at all - it’s actually about making films there, how many, well, they can film something there for years, now it can happen faster and faster, yes, that is, since well, the technology has not yet matured to the creation of full-fledged films to be honest, but the progress is such that if you’ve been doing neural networks for 3 years, we ’ve been saying that we’ve been working on neural networks for 3 years, yes, i mean it. creating a neural network that generates pictures from text. now, if you look at how neurose, the best neural network , created an image 3 years ago, then you will see that it is very, very bad, let’s watch the video, let’s watch the video, and then your questions, yeah, what can you

tell denis, comment, i will comment, this is still animation, this is not a full-fledged video, kandinsky did it, yes, kandinsky did it, as if how he does it, he draws the first frame from the text, and then you choose, that is, how does animation differ from video, in that animation is a camera flying around a static object, or what -this is the movement of the camera, and the video is a full-fledged movement of everything, here it is still animation, because there is the first frame, and then we select the camera movement, well, that is, the monotonous movement of the camera, such a zoom, and we select the first one, draw the first picture by promtumin, and we can do it without creating a new model. on based only on models that generate an image from text, generate animations like this, let's tell you where to look, yes there is, there is actually a telegram bot within which you can create animations like this, that is, everyone can do it, everyone can do it, well, that is, it copes with

complex tasks, well, that is, telegram baud - the current task was, in my opinion, very difficult, i honestly don’t remember exactly what kind of tasks were there, that is, it doesn’t require a lot of memory. everything is created in the cloud somewhere, yes, everything is created on a supercomputer, and oh, a supercomputer is an interesting topic, well, the training of the model itself and the application of the model, well, it’s called inference, it requires computing power, yeah, the model is huge after all, it’s on an iphone, roughly speaking, it won’t run on your smartphone, although progress is also being made towards this, so that it runs on a smartphone, well, at the moment kandinsky is running on the cristofar supercomputer, yeah, this is our supercomputer on which we teach models, yeah. both linguistic and these generative ones, which turn the text into picture in the video, and well, that’s where we actually use them, when you make a request, it flies to a supercomputer, it’s processed there, it’s there that kanzinsky seems to live, yeah, that’s where he makes

a picture from the text, the picture is returned to you , what do you use a telegram bot or , well, in a bot this is a certain number of iterations, that is, i go into the bot, talk about what kind of picture i want to create. which is much better, because you don’t stand in line, you do it there, but we try to keep a lot of resources on the supercomputer so, so that the queue is still satisfied, so that there are not, say, too many people in the queue at the same time, i wonder why, do you feel that competitors are spying on you, we feel it, that is, you

feel this attention on yourself, yes, but since they are spying on you, it means you are doing something that apparently causes them, i don’t know , fear, with everyone who is involved. creativity and create, competitors are spying on something, to be honest, and this is not artificial intelligence - this is not an exception, of course, but just how much you are ahead of them. this and competitive struggle, relatively speaking, you read the news in the morning there in the evening in any case, and you probably also note to yourself, did something happen that would affect who beat whom, so to speak, at the turn? this is true, of course, yes, like, say, the area of ​​​​creating large models, although it requires a huge number of computing resources, data and specialists who will actually train it all. but in principle, large big tech companies certainly have such teams, they

are developing these models, so for example, yandex has a masterpiece and the guys are making a fairly similar technology, the only thing is that they don’t have video generation, but today we only watched animation, it’s not a video, it’s still a full-fledged video - it’s the movement of everything in a row. i have a question that concerns copyright rights, intellectual property, that’s clear. what leo tolstoy, roman said in the beginning , everything is there, leo tolstoy was not passed on as an inheritance to anyone, he is an open story, yes, that’s when you load data into a supercomputer, and you license it or they kind of in the public domain, or is it not necessary, well, they’re not really in the public domain, that is, they’re more likely in the public domain on the internet, yeah, when we uploaded them, after all, it’s like we’re a complete dataset of them , we don’t post them anywhere . of course, yeah, it is stored there as a certain base, as a certain base, yes, why actually we

can’t do it any other way, because the model needs to be shown, at the training stage, it really needs to be shown pictures, well, that is, it’s like a truly educational training material for ok, related question, this is the 2800 first picture, who is the author of it, artificial intelligence or the person who did it all. generated, oh, this, this is a very good question, in fact, who owns the copyright, and , as it were, at the legislative level, all this is being developed and, let’s say, at some point it will probably be enshrined, although this is a very difficult question, you know, as they used to say, there is an angel on my shoulder, an angel was sitting and he told me, that’s when the creators of someone have something, what is the difficulty here? the matter of determining who the author is? there is data, of course. behind each photograph there is some kind of author who took this photograph, that is, there is an author of the photograph,

it’s another thing when there are one and a half billion of them, then it’s as if these authors are all somehow mixed up, and most likely there is no author, there is no author of one and a half billion photographs, that is, this is a huge number of people, there are millions of people, on the other hand there is a company that purchased a supercomputer with its own money, this is a very super expensive undertaking, in fact, all calculations are done on this supercomputer happened. well, some participation in the creation of the final picture, they made their contribution, but

so far the story is that if a person, most likely, if a person, well, logically, each company does it in its own way in the end, if a person as if he were paying for the fact that he used the neural network and received the picture, then he is still the author, that’s most likely the case, and if the motives are guessed from the motives, usually, if something is created, he still designs it? model that is not posted and is not in the public domain, kandinsky, opposite history, since our one of our first

missions, i already talked about this, is to move the community as a whole and move science forward, then we put the model in the public domain, you can use the model, not at all without some kind of payment. then the company should deploy it on its own and why create such a model that is not widely available, so that... a circle of people would use it or what not, well, to sell the generation, so they brought us on our wedding path to ryazan, a little nastasyushkety. you think it's just for you we’ll give away the bride with this christmas tree, with her beauty, we ran into the wrong people, anyuta, you’re here, dish, we need to prepare a hot dish, ram cookies, these are such ritual cookies with the desire for childbearing, fertility in the family, it smells very tasty, we’ll use two textures of fabric podskater, this will highlight

the beauty of the bride’s costume; the brighter the color, the closer the festive costume is to the age of childbearing. an elderly woman, completely black, well, let's go for ransom, why did you come to me, i'm not coming to you, hello, after all, at that time we knew how to relax, our premiere, we are playing a wedding, today is the first one, i love these resort parties, bars, restaurants, music, sun, dancing until the morning, we live on the vibrations of noise, so... we don’t even notice anymore, welcome to bodrum, there is a stereotype that turkish men are partying, this is not a national trait, all over the world there are men who like to flirt, they look after beautifully, they give gifts, well, we don’t attach importance to gifts, life itself in such a place is already a gift, so what do you have here, i recommend pistachio with white

mix, come on, damn it, take it, take it, take it, take it, disgraced, more, ayma! the lives of others are premiering today on the first day, hello, what kind of delivery is needed, i prepared the mows myself, i need to work, thank you, home! why should the plate be returned, this is the custom, and in our yard from july 7 to sunday on the first, we continue the creative

industry podcast, our guest today is a person with.

he has some kind of white canvas, well, not necessarily white, just white paper, some kind of empty, napkin, napkin, yes, there is a tablet screen, he has some kind of idea, definitely, he starts sketching out some details, with a pen, a stylus, whatever, he adds details, maybe removes something, well, a person also has patterns in his head, that is, he has memorized, he has something, here he has patterns, this is after training, step by step, he just creates a picture, how does he create a condin? model, in principle, you can draw an analogy, he begins to draw a picture from the so-called white noise, what is it, well, this is when there is no signal on the tv, there is such interference, just some kind of the matrix is ​​like this, naturally there is your request, this is the same idea, as if in a person it is expressed by these signals of neurons, in neurosi it is expressed simply by numbers,

which are the text in which the text is encoded, and from this matrix step by step in the process ... removing noise in the right direction, an image is created, that is, creating an image from this white, this white noise, from this matrix, you can simply visualize, that is, there are 100 steps or 1.00 steps, for which from any of this noise is created and visualized, the final picture appears, so to speak, for example, you want to make a pug in space, yeah, you always have white noise at the beginning, and step by step you get a pug in space at the end of this process. if the noise is a little different, it is clear that the noise can be whatever you want, swap two pixels there, it will be different, but also in space, this matrix is, in a sense , responsible for the imagination of the model. this can be visualized, but this process, it’s called denoising or denoising, just so that it happens correctly, here

we need to show this neural network one and a half billion picture-text pairs, that is, so that it learns to connect text and image, in principle, there is an analogy with a person in this sense, and you do all this every day, and we do all this, well, we do all this every day, everything this is what we do, and we achieve excellence every day.

put, in general, this is also done by neural networks, others, they do not make a picture based on the text, but based on these time series of brain signals they make a picture, the same one, in general, as if you were connected to this if you install such an electrode, you can actually train a model that understands you without words and draws a picture if you need to draw a picture, but the limitation of these models is that for each person, for now, you still need to teach your own model. that is , for each person, when he imagines a palm tree, the signals are slightly different, that is , it is still impossible to train one model for all people, and

it is expensive to train a model for each person, and you need a supercomputer for each person, you need a supercomputer, until there is some kind of universal thing that everyone understands, in general, i hope, i hope that not soon, and moreover, it will still remain, that is , there is no need to be afraid of it, even though it sounds futuristic somehow, but there is no need to be afraid of this either, because in general.

who is heavier than a kilogram of fluff or a kilogram of lead, well, it is clear that the neural network saw in the training data the word fluff with the word light, it saw much more often than fluff with the word heavy, well, on the contrary, which is logical, and of course it answers that a kilogram of lead means heavier, a kilogram of fluff is lighter, uh-huh, and explains why, well, this is fluff for perfection, there are three three paragraphs there, why it is really dangerous, because a schoolchild, for example, who looks... here is the answer to this model, he will believe her, then he will go on to broadcast this thought further to his peers, and so on, and what to do, we need to fight hallucinations, in principle, this is the area that is now

most actively developing, like this large language model, generative model, teach us to hallucinate less, well, we started with the fact that we, in general, are lukewarm , so to speak, if we had created... our digital images created by artificial intelligence, it would be interesting to look at them, of course, here are our models and here lena was thinking about what we could look like, we just have a question about this, oh, yoshkin, who is this yoshkin’s cat, it’s okay, it’s some kind of guy, you know how pardinsky sees us like that, yeah, yeah, wow , and why did he make us look so young? well, yes, he’s hallucinating, and i, you see, in the form promta in my head, you look exactly like that, and i’ll actually tell you how

- we get photorealistic... portraits, we have a team that deals with detecting fakes, you know, now this is especially relevant, a pressing problem, well, with the development of generative models, you can generally make a bunch of fakes and post them on the internet, and this can bring huge losses to companies and states and all sorts of reputational costs, that is, you need to learn to distinguish what is created by a neural network, what no, actually, as part of fake detection, we decided to make a model that generates fakes, why, then, so that on that dataset, which...

has already helped the fake detection model, it simply transfers the face, that is, this is the real you, in in principle, it is, yes, dear friends, denis, dimitrov, our guest in the creative industry podcast on channel one is the man who created the kandinsky neural network of a wonderful team, and the main conclusion, even if we are not quite everything, in general in the end we understand, there’s nothing to be afraid of after all, look, what nice, smart people who do those things.

first channel. see you. dear graduates, we congratulate you on your graduation from school. now you can behave badly. but the most important thing is that your dreams become goals. and you, never doubting yourself, rushed to achieve your plans.

i will come for you tomorrow after seven, i am the blue sun, i appear less often, less often, it hurts me so much to watch how beautifully you lie, you are on the body of the river, you feed on me, because my kiss, like a knife, i cut you, earth, kings

of me . snow, love will save world. sleep, tomorrow i will rise for you with the first ray, i want to be weaker, i want to be with you more and more often and closer. i will burn you, but i will already

crush you with my hot shoulder. to the white snowy lips and i will never, never see again, hey, earth, flood me. beyond the snow, love will save the world.

hello everyone, this is a podcast 20 years later, i’m its host konstantin mikhailov, we have a special episode dedicated to the graduates’ holiday, mine helps me. hello friends, hello, today we will have, one might say, a marathon, we will have guests come, go, replace each other, meet, see off, talk, talk, and the guests, of course, will sing, so stay, by the way, friends, for the first time for the first time

, i thought that no, for the first time this holiday was celebrated in leningrad in 1968 year, in fact, like the main character of ala prusa’s book, students or graduates were entering a new life, so it was... a decision was made to call this festival exactly that, then it was closed for some time, and then in 2005, in my opinion, he was revived, and now he is rushing at full speed towards his fate, he said in aloud parya, meanwhile, friends, nike borzov came to visit us today, the man on whose songs i grew up, and we grew up, this girl, a pearl, yuliana koraulova grew up on my broadcasts, great, hello, yes, i ’m actually sitting with the legends today. and i feel simply incredible, thank you very much, and so do i, nike, please tell me how your graduation went, do you remember it? yes, it went well, uh, well, after, of course, the official part, where everyone was handed out diplomas and so on, there was an unofficial part, it

was very fun, i remember that our graduation ended in the morning - in sukhanovo on the lake, we took several boats, sailed to the middle and just lay and watched and looked at the stars. i still have this pleasant feeling, this is the moment i remember most, it was just beautiful, friends, we listen to the song, and after that we meet again in our studio, because galibri and mavik will come to visit us. congratulations to all graduates on graduating from school, i wish you a wonderful, happy journey, and of course, remain riding the star. clinging to the rays, with the moon on a leash, in the night, dangling from the stars, rushing towards

the wind, an unbroken dream, a claim! oh, life is wonderful, and life is wonderful, quite, you can be a little. it’s dangerous, take my heart, don’t remember me, believe me that everything is wrong, the dust of a tear over the forest river, lost forever. peace, feathers on

the star, don't pick up the apple, one way ticket, spring, you are very beautiful, you are very beautiful, you are very beautiful. quite, you can be a little dangerous, oh, take mine heart, take my heart, don’t think about me, believe me that everything is not in vain.

riding a star. oh, take my heart, don’t think about me,

believe me that i haven’t typed everything. permmu. wow, again adventures, how beautiful you look, it was here that ivan the terrible founded astrakhon, there is a legend that stepan razin on the banks of our astrakhan ravines kept his treasures in the barrels of cannons, all the cannons were found, all the barrels were found, please tell me, the search continues, this is one of my best things in life, what will you give us? unusual borscht, fish borscht, i would come back for

this borscht more than once, have you decided what you will cook? yes, remember, gita told us about caviar, i’ll marinate it, add a secret to it, and i’ll make cold soup, then i’ll add smoked fish, it’s time to start, let’s go, premiere, cooks on wheels, today is the first one, asl didn’t like living in the common apartment with a goat heifer, so he made a firm man’s decision to move in with the bull, in the cartoons we are yet...

it’s not your fault, i just didn’t immediately understand what i really needed, subtle matters, today on the first. is it that the general didn’t splurge on the bouquet? he was generous, the bouquet was at the dacha, it was from me. at the entrance to the legendary bamov village, we are greeted with a starry look by this wedge heel. 50 years ago,

the pioneers got here on it, and they landed there on the banks of the tayura river. then there was no bridge, no roads, just continuous, impassable taiga. i. we didn’t look at the frost at 50 and 53° in the rain under all conditions, but when i reach the horizon, we’ll see what the horizon is, but the horizon goes away and you’re chasing after it, a group of 13 of us arrived, a year later there were five of these 13 left, a year later there were two of us left of these 13, that is, to get used to this road, to get used to the routine, to the slopes, i saw what it was like...

alexander joined us mavik, hello, hello, and evgeniy galibri, hello, zhenya, hello, sasha, hello, sasha, well , the beautiful yuliana karaulova is staying with us, yuliana, anton, please tell me, here’s alla and porusa, you performed at this magnificent holiday , yes, we did, last year, did the scale strike you, tell me, did the scale strike you, in general, to be honest, i wanted to perform at the scarlet prussas for so long, but for some reason it never happened, although i have been singing... for about 100-500 years, for the first time last year i was called, i can say that for it was legendary for me, there were a lot of people, in short, it was cool, the whole square sang with me, but the coolest thing was actually after, because at night i went for a walk along nevsky prospekt, and i was just shocked how many

people there were at 3:00 a.m., white nights, the city is blocked off, and after a sailboat enters the water area, it’s just all... exactly the brightest of these moments, were there, my dear, wives, sasha, do you have stories when , perhaps, something went wrong at graduation, or vice versa, everything went so that you were like, damn, i would like to be it’s also fun to celebrate among them now, in fact , i was impressed when we performed the year before last, this scale and 110,000 people, in my opinion, were on the square and when you

’re standing on stage and you just don’t even see where it all ends, where the studio ends ? it really struck me, inspired me, well, yes, we usually don’t do that much. we are worried about our performances, but then it was exciting, however, there were a lot of people, there is such an interesting scene there, yes, and by the way, you know what i also want to say that the scarlet russians in st. petersburg are not the only ones, because there is also a cool graduation in vladivostok, i performed there probably six years ago, a sailboat also comes out into the water area, also with scarlet sails, it is not called scarlet sails, but in fact, it repeats the action, class. what did she sing? and sang straight through the entire program. maybe i should ask you to sing? well, ask, sing, please okay. friends, as you probably guessed, now yuliana koraulova will perform her amazing song out of orbit. there is an amazing, bright life ahead of you.