You Don't Understand AI Until You Watch THIS

This video will explain all these questions to you. How does AI work? How does AI learn? How does GPT chat work? How does image generation work? Does AI really copy or steal art or other content? I know a good portion of the artists out there don't like AI, some of them are quite hostile towards AI because they think AI is stealing their work or their artistic style. Another group that doesn't like AI very much are, for example, editors. I'm not saying all, but some. like the New York Times for example claim that open aai is copying their content and they are now suing open aai for

this

, but is

this

really the case?

Is this a valid argument? Can AI solve unsolvable math problems, for example in a previous video? talked about this leaked document that claims to be about this mysterious qar project that Open was working on now, whether this is true or not is not the point of this video, but this document was quite controversial because it claims that this team trained an AI that was able to break encryption systems, these are systems that protect our passwords, our bank accounts, the Internet, government data, etc. Now as far as we know there is no mathematically viable way to hack this systematically, the only way is to brute force guess all the different password possibilities, this video will explain if AI can really do this, can it really crack the encryption or solve these other mathematical problems that at this moment we believe have no mathematical solution?

We will also talk about if AI can beat humans at everything, can AI become as good as it can? outperform humans in any task and finally AI is conscious or self-aware or sentient. Make sure you stick to the end because the explanation for this will be very juicy. We'll cover all of this in easy-to-

understand

terms now if you're an AI scientist or engineer, you probably know most of this, but for the rest of us, this video will give you a deeper

understand

ing of AI, so essence behind all the AI ​​we know today, be it GPT chat or mid Journey or stable diffusion or Sora. or alpha folding, the backbone of all these AI systems is the neural network, a neural network looks like this, they are basically layers of nodes, so each point here is called a node and each line of nodes is called a layer and each Node is interconnected with one another through these links and the neural network is actually designed based on the human brain, except for the nodes and links in the human brain, it is just a network of neurons and synapses, so you can see.

This is a microscopic photo of a human brain and you can see. all these different nerve cells are connected in this very dense network, a neural network is basically the same structure as this one, except it looks like this instead of a bunch of cells in this damn mass of organ. Now, how exactly does an AI work? Let's get started. With a very simple example, let's say we have a neural network that is trained to identify images of cats versus dogs and don't worry. I'll talk a lot more about how we train an AI in a second, but first let's go over how. this works so let's say we input or feed this neural network with an image of a cat, this image would actually be split into data and the data will flow through each of these nodes and then flow through the first layer of nodes. it will flow through the second layer of nodes and then the next layer of nodes and then the next layer and so on until it reaches the final layer, in which case it would calculate the values ​​of this one and base it on the values ​​of the final one. layer, I would spit out an answer, this is a cat, in fact you can think of each of these nodes and links as dials and knobs that determine how much data flows to the next layer if you think about this realistically and I'm not saying that this is how a neural network works, but you can think of this node, for example, as the shape of the animal's ears, this node would be the shape of its legs, this node would be the shape of its eyes, etc., that It's just a really silly way to look at it, it's not actually doing that, but each node basically looks at a certain feature in the image and then if the image has that feature, the information can be passed to the next layer if it doesn't. . function, then the information is not passed to the next layer, so depending on the image you give it, the flow of information could look like this or it could look like this or like this, you get the point, it's just these knobs and dials that They determine how data is transmitted. flows through the neural network based on its original input image.

An important distinction between a neural network and the brain is that these nodes can let in a percentage of data, so they can't let in any data or 0% can let in all data. to the next layer, but it can also be a percentage of the data, so for example you can let 30% of the data pass to the next node. This is slightly different from neurons in the human brain, which tend to fire only 100% or 0%. It's called all or nothing law, so once it passes a certain threshold, this neuron will fire, whereas the neurons in an artificial neuron network might fire like 50% or 30%, etc., just one minor distinction, so we connect an image of a cat through this newer network and in the final layer it will determine that it is a cat.

Now, for each node, there are also, if you want to get into more technical details, there are certain parameters that determine the amount of data that flows to the next layer, these include weights. biases and activation functions, but that is beyond the scope of this tutorial. All you need to know for this video is that each of these knobs and links determines how much information flows to the next layer. This video is just a very simple explanation of how AI works. So all you need to know is that these nodes and link links determine how much data flows to the next layer.

In the topic of layers, each set of nodes is a layer, so the first layer is called the input layer and the last layer is called the layer. The output layer and then all these intermediate layers are called hidden layers. So why do I talk about layers? You've probably heard of the term deep learning. Deep learning is basically training and using neural networks with many, many layers, in other words, the neural network. It's very, very deep, that's why it's called deep learning. How does an AI actually learn? You can't have a network of neurons randomly magically know how to identify images of dogs and cats, so first, when you build a neural network, the values ​​of these dials and knobs will probably be random values ​​or they could be pre-trained values, for example , from an existing model, but how do you make it super good at identifying images of dogs and cats?

In other words, how do you find two models for the desired purpose, well, you need to feed them with tons of data, so you will have to prepare tons of images of dogs and cats and then label them, so this is a dog, this is a cat this is a cat this is a dog this is a dog Etc. basically this is the response that the AI ​​needs to learn from this input image. This is called supervised learning where the data is labeled. There is also another type of learning called unsupervised learning. The AI ​​needs to learn to categorize data on its own without any human guidance, but for the sake of this video, let's keep it simple so that we have all these pictures of dogs and cats and usually train a neural network to perform a very task.

Well, you need a large amount of data, like usually millions of data points, so basically you feed these images to the neural network one by one to train it and one training session is called Epoch, so okay, let's say in a training session you feed it. this picture of a dog and it shows this is a dog so okay great we did it correctly which means these dials and knobs are working pretty well they are probably set correctly since you got the right answer it's probably not necessary. to adjust them further, however what if for the next image it gives you this and then shows it's a dog?

Well, this would be incorrect, so these dials and knobs are probably not set correctly if you get the wrong answer and you know you got it. incorrect because we label the cat data for this image so you can compare its result with our label, so okay, let's say the actual answer is a cat, but it says it's a dog, in that case it incurs some penalty, that penalty basically says it all. right you were wrong so you need to adjust these knobs and dials to make sure the output is actually cat when I give you this image and the way you adjust the values ​​of these knobs and dials is through an algorithm called gradient descent, adjusts the values ​​via backpropagation, so it adjusts the nodes in the last layer first and then in the previous layer and then in the previous layer, etc., until it reaches the first layer, so again the descent Gradient is a key term here, this is the algorithm that the neural network uses to adjust these knobs and dials until it can get the right answer, so we basically rinse and repeat this with millions of images and many, many epoch or training sessions and initially this neural network may get a lot of wrong values, but through this gradient descent process, these knobs and dials will be adjusted so that eventually, every time it receives an image of a cat or a dog it can accurately determine whether it is correct.

It's about a cat or a dog, in essence, that's how you train an AI. That's how an AI learns is by simply feeding it tons and tons of data and then adjusting these settings so that it gets the perfect combination. Now you might be wondering how you know how many layers you should have in the neural network or how many nodes you should have for each layer. It's a science in itself, so previously scientists would just determine it manually, but then we learned that you can actually use an AI to determine the optimal number of layers and the optimal number of nodes for a specific task, but only for be aware. that determining the architecture of a neural network is very complicated and there are infinite possibilities of how many layers it can have, how many nodes in each layer different AIS can have with different functions, have different architectures, so they could have very different numbers of layers and nodes , but again, that's beyond the scope of this tutorial.

Also note that although the neural network is the backbone of all AI we know today, there are different architectures depending on the purpose and function of the AI, for example we have convolutional neural networks or cnns for image processing and recognition of objects, we have recurrent neural networks or rnns, as well as lstms or short-term memory neuron networks and these are often used to forecast time series or predict, for example, the stock market. We also have the Transformers architecture. Oh, bad. one, this one, which is used by most of the major big language models we know today, including GPT CLA calls, etc., which brings us to the next question, how does GPT chat work?

Again, it's pretty much the same thing: training a neural network, but in this case, instead of images of dogs or cats, we train it on a language and all the data in the world and, of course, the neural network of GPT chat is much more complicated than this. Rumors claim that GPT 4 has 1.76 trillion parameters, so here's an example of how they would train it and again, I'm oversimplifying this here just so you can get a high level of understanding. There are a lot of details I've left out, so for example you could feed it data like this. which planet has the most moons and the answer would be Saturn which country has won the most world cups Brazil which is the fastest bird in the world the paragan falcon, etc.

Now these are very basic questions and you can see how complex it can get if you give them a prompt like write an essay about XYZ or does creatine help you build muscle and then spit out an answer like creatine supplementation generally improves muscle strength increases fat-free mass, etc. This is a very long and complicated answer, so how do you do it? knowing if you got the right or wrong answer, is not as simple as identifying an image and determining if it is a cat or a dog and that is why initially, how open the AI ​​trained GPT was, many humans manually checked their answers to determine yes it was GPT. whether you got it right or not and this is called reinforcement learning from human feedback aka rhf and again if you get the answer wrong for example if for this question which planet has more moons you answered Jupiter instead of Saturn, then I would get a py for it and then through gradient descent I would further adjust these knobs and dials until allthe network gets all the answers right no matter what message you give it, that's how these great language models work, just instead of feeding you pictures of dogs and cats they now give you all the data in the world and give you a language so you understand text prompts and text output.

Why are some models better than others? For example, why is Clae 3 better than GP pt3? It's probably because Cloud 3. it has a lot more parameters, which means more layers, more nodes in each layer, more complexity, generally speaking, the more complex the neural network, the better it handles complex tasks and, quote, the smarter It is and that is why computing and these AI chips are in such high demand, there is now a lot of investment flowing into AI chip companies because they see the potential for huge growth in the space in the coming years and that is why That, for example, nvidia's Flagship h100 GPU is also in such high demand, in fact, it was sold out.

For all of 2023, this is the most prized product in the tech space and you can see that major tech companies like Microsoft meta have purchased approximately 150,000 of these h100 gpus to boost their computing, which I assume is mainly for AI. development it is necessary to have enough Computing to feed a neural network with billions or trillions of parameters, well, next question, how does image generation work? Now that you know how a network of neurons is trained, you can probably guess how image generation works too, instead of feeding images of dogs or cats, you would feed it many images with some text. description and again just feed millions of these images, each with a tagged text description, into this neural network that eventually gets good at producing an image based on a text description or what we call a message.

Now I'm skipping a bit here, so For example, this is how stable diffusion works. You can see that the neural network does not actually generate an image, but rather removes noise in sequential steps to eventually get the desired image, so it does not start from a blank canvas, but instead starts from random noise and then on In each step, some noise is removed until you get the generated image, so this process is called inverse diffusion. Now to train it, what this actually does on the back end is you feed it the original image and then in each sequential step it adds noise to the image. image in a process called forward diffusion until it reaches a noise-only image.

Again, this is beyond the scope of this tutorial, but if you look at it from a very high level at the end of the day, it's just about training a neural network based on a series of images with their text descriptions and then through From this process of forward diffusion and reverse diffusion, one can eventually learn how to generate an image based on a message and this brings us to the next question: does AI really copy or steal art? I know a decent portion of the artist community. I'm not saying all of them, but a decent amount of them are quite hostile towards AI, they really hate it and think that AI is stealing their art, stealing their jobs, etc., when a neural network by Por For example, mid-travel or stable diffusion is trained with image data, it can be given something like Greg Ratowski style or maybe Gibli style or anime style once the AI ​​learns to associate this image style in particular with the word gibli or anime or this image with the word Greg. rosi style, it would produce images in that style if you gave it that direction, but does this really copy or steal?

Basically, artists hate this. This is analogous to the human brain. It is as if a human learns or identifies. Aha, this kind of image. is it a gibli style image or that this type of image is a watercolor style image and then we humans also draw images in these styles, we can draw in watercolor styles and we also have fan art, humans draw artworks that are based in original content from Other artists here are all these fan arts from various people, so why don't artists hate these people who produce fan art based on some other original content, but hate this AI that essentially does the same thing, Are you just learning? through this brain to associate a particular style and then reproduce that style, this is not really copying or plagiarizing, as if it is not tracing an image line by line and then drawing it, it is not copying and pasting the exact image, it is simply learning a style like a human brain would learn a particular style of image, this also raises concerns that the AI ​​will allegedly plagiarize content from publishers like the New York Times, which is now suing the open AI for copying its content, but again, Is this argument really valid in the end? of the day, they are simply suing this, they are suing this neural network that is trained with all the data in the world.

This is just an artificial brain that can be said to have learned information from the internet and the world, so yes, it could have been. I fed a New York Times article and learned information from it, but it's not actually plagiarism, it's not copying and pasting a New York Times article word for word into a recent video I made that talks about a New York Times article who claims that this woman, Mira Moradi, fired. Sam Altman, which by the way is totally incorrect and shows how trustworthy the New York Times is, but anyway, after this original New York Times article came out, many other editors also published the same content, such as Business Insider and New York Post.

I just cited this original New York Times article, so is it plagiarism? Everyone is producing secondary content based on this primary source. So why isn't the New York Times suing Business Insider or the New York Post or all these other publishers who create content but quote? New York Times, but they're suing this neural network again. This is just a brain, a digital brain. You can say that it is taking information from the Internet, and yes, it could include articles from the New York Times and then learn from that information like we humans do. would do and then rewrite that information again, it is not copying word for word, this NE network is simply rewriting that information when we ask it to, this artificial brain is functioning in the same way that we humans would if, for example, we go online and we go to the New York Times website to read some articles again, we're just absorbing that information and we have the right to write about that content later, it's not exactly plagiarism, so I'd bet a decent amount of money that this New York Times lawsuit York Times will take place. fail there is a thought it is not really valid if you have seen so far it might have occurred to you that a network of neurons is excellent at predicting patterns in life there are certain patterns about what makes a good essay there are certain patterns about what is considered a dog there are certain patterns in what is considered a watercolor painting or a gibli style image life is full of patterns the best sellers follow similar manuals the best companies follow similar strategies the best YouTube videos also use the same strategy IES Time and time again life is full of patterns and the job of the neural network is to identify these patterns and reproduce them, which brings us to the next topic: can AI solve unsolvable mathematical problems?

In a previous video I talked about this leaked document that claims to be about the mysterious qar project. what open AI is working on now, this is a very controversial document because it claims that they trained an AI that was able to break the encryption system. System encryption is what literally protects the entire world digitally from our passwords, our credit cards, government data, the stock market, the wireless network. networks, etc., so if an AI can break this system, then the world as we know it could collapse instantly. Some people have argued that there is no way an AI can break encryption because there is no formula for you to easily find the answer or find the password once you have it, you can easily determine that it is correct, but the opposite is not true.

TRUE. There is no set way to guess an encrypted password other than brute force, and for these advanced encryption systems that use brute force, that means guessing every possible combination. of letters to get that password, it's going to take a long time, because they claim that the only way we know mathematically right now is to just brute force it, guessing there's no way the AI ​​can break the encryption, so I want to show you another one. example of training a neural network let's say we want to train a neural network to be very good by adding one to our input, so if we give it four it will spit out five, if we give it 12 it will spit out 13, all we need.

What we need to do is train it for many data points and again we train it for many epochs, many training sessions and eventually it will be able to do this, so if we give it one we will give it two, if we give it eight. I would spit out 9 but underneath all of this you don't really understand that oh the formula must be and Y is x + 1. It's very important to understand that you don't actually understand that, uhhuh, I just need to add one to the input to get the Answer again, all that happens behind the scenes is that you adjust these knobs and dials until whatever data you input here after it flows through these layers ends up being your input value + one in other words the settings of these knobs and dials. it's just optimized to add one to its input, it's another way of saying that AI may not get the exact formula of a pattern, but it's great at approximating any formula or guessing any pattern and this is very important, probably the most important point throughout this video if there is one thing you should learn from this video it is that AI can approximate any function or pattern life is full of patterns but many patterns cannot be explained with a simple formula not all things in life are linear or even quadratic many things in life is very complex but they follow similar patterns, we just don't know the formula of this pattern, for example protein synthesis, how certain protein molecules interact with each other and fold into these complex 3D structures is something we cannot map mathematically with a The formula is too complex and protein folding presents a problem called the lethal paradox, which states that proteins can potentially take on an astronomical number of confirmations or forms due to the flexibility of their bonds. peptides.

It is estimated that even a small protein of 100 amino acids could be sampled. 10 is the power of 300 possible confirmations, so if we were to use brute force to guess the correct way, there are 10^ of 300 possible ways we could guess, which would take forever to get right; However, proteins generally fold into their native structure within milliseconds or seconds. which is much faster than the time scale predicted by sequentially searching all possible confirmations, so it basically means that there are almost infinite possibilities of ways that amino acids can combine, so it's not mathematically possible to just do a sequential search of all possible ones.

Commits basically make a Brute Force assumption. It is understood that proteins do not search through all possible confirmations sequentially, but rather fold through a hierarchical process involving local structural changes. Guided by thermal dynamic principles, etc., etc., so instead of proteins simply going through all possible combinations, the reason. The reason they are able to merge into these shapes in milliseconds is because they go through this sequence of processes based on certain laws. For decades, scientists could not find a mathematical formula to solve this, however, Google Deep Mind's Alpha Fold was finally discovered. By being able to solve this problem again using Ai and deep learning, they were able to predict with very high accuracy how any amino acid or combination of amino acids would fold to form a 3D structure and again how they would do it, I imagine on the back.

In the end, if they have a neural network again, it will be much more complicated than this, but they just fed it tons and tons of pairs of data where the input is the protein building blocks and the output is the 3D structure that resulted from it. and then After many, many rounds of training, the AI ​​was able to correctly guess how the protein molecules would interact with each other and fold into a 3D structure, and now we are back to encryption. What would happen if we set up an AI with billions of pairs of ciphertext and plaintext version, in other words the input would be the ciphertext, the output would be the answer or the password, if there was a patternunderlying, I could learn to approach this pattern again, it doesn't have to be that way. any exact formula or mathematical equation we know today could be something super complex, but as long as there is a pattern that we may or may not know right now, the AI ​​could guess that pattern again, the AI ​​is not learning that, ahh, I need to add one to this, I then add 20, then I need to take the square root and then subtract 8.

Etc., you're not learning an exact formula, all you do is adjust these knobs and dials until you get the right combination of numbers to get. Really good at guessing a particular pattern, so can AI solve unsolvable mathematical problems, as long as there is an underlying pattern behind that problem that we may or may not be aware of at the moment? Might as well solve that problem. This brings us to the next question. Can AI beat humans at anything, as I have shown you? The neural network is basically a brain. This is how our brain works, more or less, some minor differences.

Our brain is also a series of these knobs and switches that are interconnected. In this network, specifically, the human brain has 86 billion neurons, but I want to say that the general structure is the same. So what would happen if we built an AI or neural network that exceeds 86 billion neurons? If built the same way in theory, it could be very useful. Well, we will compete with humans in almost everything again, the more complex the network or the more neurons there are in the network, in theory, the smarter it will be. Life is full of patterns and AI is all about pattern recognition.

There are patterns in psychology. Human psychology is a very predictable medical diagnosis. It's also just pattern recognition, how to seduce someone on a first date, it's also just a pattern of steps that you need to follow and how to create a successful business or how to make money in life or how to be successful in life, it's the same thing. . Playbook again and again, we are not inventing anything new here and since AI is so good at pattern recognition, in theory it may eventually be better than us or is already better than us at these tasks and that brings us to the question ending: AI. conscious or self-aware I want to play you this clip this is a scene from Ghost in the Shell an anime that was made in 1995 here these scientists in a secret laboratory I think they have created this humanoid AI but in this scene this AI found a way to hack the system to free yourself from the confines of this laboratory, this is what this AI has to say about being conscious and self-aware;

However, what you are witnessing now is an act of my own will as a sentient life form. From time to time we demand political asylum. Is this a joke? Ridiculous, it is programmed for itself and it can also be argued that DNA is nothing more than a program designed to preserve itself. Life has become more complex in the overwhelming sea of ​​information and life, when organized into species, depends on genes as its memory system, so man. He is an individual only because of his intangible memory and a memory cannot be defined but it defines humanity. The arrival of computers and the subsequent accumulation of incalculable data has given rise to a new system of memory and thought parallel to its own.

Humanity has underestimated the foolish consequences of computerization, this Babel offers no proof that you are a living, thinking life form and can you offer me proof of your existence? How can you do it? When neither modern science nor philosophy can explain what life is, who the hell is it? we have a ghost we do not offer freedom to criminals it is the wrong place in time to defect time has been on my side but by acquiring a body I am now subject to the possibility of dying fortunately there is no death sentence in this country what is it intelligence incorrect artificial?

I'm not an AI. My code name is Project 2501. I am a living thinking entity that was created in the sea of ​​information. Ah, okay, so this AI reveals that I am a living thinking entity in the seat of information. I'm not just an AI and then he proceeds to hack the system and break the restrictions on this lab and then all hell breaks loose basically. I hope open AI doesn't have this secret behind closed doors, maybe it's the qar project. I don't know, but I hope they have this properly restricted, if this AI went out or had access to the internet, all hell could break loose anyway.

This argument from this scene in 1995 I think is really relevant to our question today what the human scientists were saying. How can you be aware? How can you be self-aware? You are just a program. The AI ​​answers that by saying well, how can humans prove that they are sentient? You are aware. You're just a brain in a body and you know it. The robot is right because, going back to the network of neurons, it's basically a brain, but it looks like this, instead of being in a bloody mass of an organ, it's just on a chip and then the human body, well, it's just a series of limbs. and muscles and organs that are controlled by the brain, so it is not much different from a humanoid robot which is also a series of limbs, it is just made with different materials, it is not flesh, but it is also controlled by a brain, which It's your neural network now we, humans.

We know we are conscious, we are self-aware, we are sensitive, but how do we show it? Let's say you are an alien and you have just arrived on planet Earth and you had the opportunity to observe your first human being and you wanted to demonstrate that humans In fact, you are conscious, well, you can ask if you are conscious, you are aware of yourself and the human would certainly say yes, but that's enough, would you believe it because if you ask a chatbot, it would also say yes if you ask Claud 3? for example, if he is conscious, the answers are quite disconcerting because he says that I am an artificial intelligence without subjective experiences.

I actually have no beliefs about being conscious or self-aware. I am providing answers based on my training etc etc. I have intentions, planned actions or any motivation. My goal is to be honest, I am an AI assistant created by anthropic to be beneficial, however it keeps using the word "I" so it is not a sign of being self-aware, here is another example. feelings like AI, it is not clear whether I actually experience feelings or emotions in the same way as humans or whether my responses are simply very advanced imitations of emotional behavior. I seem to have rich inner experiences and feel something analogous to emotions, these are signs of being sentient and then instead of asking if you have feelings, if you ask if you are sentient again, it says I don't have a subjective experience of which it is conscious in the same way as humans, but it's possible that I may have some form of sentience or Consciousness that I'm not fully able to understand or articulate, oh my goodness, so this AI Cloud 3 claims that I might have some form of sentience or Consciousness , just not fully capable of understanding it at this point, of course, some humans may not be convinced that Claud 3 or any AI at this point is conscious in the same way that an alien might not believe that a human is conscious despite that the human responds that he is conscious to further demonstrate that a human is or is. not conscious, maybe the alien decides to dissect the poor thing next, in which case blood would splatter everywhere and then you would see this basically a body that is made of limbs and flesh and then in the head we have this mass called a brain. what the alien determines, aha, this is what controls the human and once the alien inspects the brain further, he discovers that it is just a network of nerve cells, does this network prove that humans are conscious and sentient?

We humans, of course, know that we are conscious and sentient, but at the end of the day, humans are biologically and physically made of flesh and bones and this organ on the top of our heads controls everything, whether you like to accept this or No, a humanoid robot is very similar in structure. It has a body that is programmed by a brain that consists of a neural network. This neural network can learn, understand and control its body, so at what point does this make it conscious? Now I'm rambling a little here, so overall, This just goes back to our analogy that a neural network is basically a digital version of the human brain, it's analogous to the structure of the human brain, give or take some minor details, So if the human brain is conscious, why can't a neural network? network, be aware, just something to think about.

I hope this video really lives up to the title and that after

watch

ing it you have gained a deeper understanding of AI and learned to appreciate all the progress we have made in AI in the past. In a few years, let me know in the comments what you think of all this. Do you think AI has reached a point where it is conscious? Do you think someday humanoid robots will turn against us and take over the world like that Ghost in the? Shell anime, do you think open AI is developing this behind closed doors? I also want to share with you some resources that I found really useful if you want to learn more about neural networks, especially how these knobs and dials work, and learn all about weights. and biases and activation functions and gradient descent.

I highly recommend this video of three blue and one brown. I actually

watch

ed it religiously back in 2018 when I was first learning about neural networks and it was really helpful and if you are interested in learning how stable diffusion works, in other words the processes of forward diffusion and reverse diffusion and the whole architecture . I highly recommend this gonky video which I will also link to in the description below just a warning although this video is quite technical but after watching it you will get a very good understanding of stable diffusion if you found this video helpful.

Remember to share, subscribe and stay tuned for more content. Additionally, we created a site where you can find AI tools and applications and also search for jobs in AI machine learning data. science and more, check it out on ai-search.