Name: Brainstorm AI 2022: Silicon Valley’s Newest Gold Rush
Uploaded: 2023-12-18
Duration: 14 min 44 s
Description: Emad Mostaque, Chief Executive Officer, Stability.Ai

Emad Mostaque, Chief Executive Officer, Stability.Ai

Brainstorm AI 2022: Silicon Valley’s Newest Gold Rush

Uncategorized

Brainstorm A.I. 2022: Generative A.I.: Silicon Valley’s Newest Gold Rush

Welcome back. I hope you all had a chance to stretch your legs and grab a coffee. And now let's continue with our main stage conversations. There is probably no hotter topic in A I these days than generative A I. It's an umbrella term for software that instead of just analyzing existing data can create original blog posts, Wikipedia entries, artistic images, video code, and more up next, a demo of the remarkable tech from Stability A I followed by a conversation with the firm's founder, Ahmad Mustaq. He'll lay out the economic and creative upside that will follow from making this open source free to use code available to anyone who wants it. Please welcome to the brainstorm A I stage, Ahmad Mustaq founder and CEO of Stability A I. We'll get to the demo in just a minute. But first of all, I want to ask you, why did you do this? And especially why did you do this as an open source project? I mean, potentially, you know, this could make a lot of money in the film entertainment industry. Why do it open source, fun and profit? I think I saw this technology and this technology classical A I was you built an air for one thing, this is there that can do anything. And I realized it would be like servers and databases which are almost all open source and it's all about scale and service customization. So I thought rather than wait five years to get there, let's just go there at once and allow it to spread everywhere which it has. So like for example, stable diffusion has just overtaken Bitcoin and Ethereum on github stars in terms of developer popularity, the fastest growing open source software ever with the biggest community building. The craziest things. Wow. Um Now, one of the things when stability of 1.0 came out, that was a little bit controversial is that unlike some of your competitors like dolly, you didn't apply any filters to either the training data or the output to make sure that people couldn't generate pornography or other sort of not safe for work images. Now, with the 2.0 version that was just debuted last week, I believe you, you've actually rolled that back a bit. You have applied some filters that make it much harder to generate pornography. Why the reversal on that? So the first part was actually a research initiative led by the LMU team at the University of Munich with collaboration from ourselves runway ML and some others that providing the main compute infrastructure. And Robin Rombach, our lead GS of code are doing it and they said, let's start it as a base and make it just an image of the internet and see what we can do and that's what came out. But we put a safety filter on it and an ethical use license as well to limit the amount of output for our own release of stability. We decided and listened to the feedback that we would filter out, not safe for work elements so more people could use it. We decided that we would also make it a more open dataset. So we could really understand how the model worked and we millions of dollars kind of doing that as well, which is now results in the model that anyone can use in a commercial and corporate setting. Great. Well, let's let's, let's see the model and I this is a special demonstration. This is the first time I think you've shown this particular version of the model. Can you, can you explain what you're going to be showing us? Sure. So basically, um we released 2.0 12 days ago, we released the original 1 August 23rd, released 2.1 tomorrow. Um This is the pace of kind of innovation here. 2.1 is very special because when we released 1.01 0.4 actually in August 23rd, it was 5.6 seconds for a generation. We release that release that to 0.9 seconds for an image. Um just a few days ago as of next week, it will be 30 images in a second. So we designed it to be fast and now we've got to real time. This unfortunately isn't quite real time. Um But we've got the one second generation and again, it can generate anything. So you take words, you put it in and for example, is on stage screen. Yes, you can do a lovely Labrador, right? Labradoodle. We like doodles, Labradoodle. And then in a second, it generates a picture of multiple lovely labradoodles with a hat and then boom, it generates it with a hat. I think it's just loading on the internet and then we can do by Van Gogh and it becomes by Van Gogh. Um So this is actually a very interesting thing because 100,000 gigabytes of images compressed into a 1.6 gigabyte file that works on your Macbook. And as of last week, it works natively. So you'll be able to generate these images in a second in a stained glass window. So you can generate just about anything because it understands 2 billion different concepts. And this is just unimaginable, right? And like I said, as of next week, it will be 30 frames. A 2nd 2.1 adds back a lot of the artistic elements. We got some lovely stained glass windows. Isn't that nice? But then it's also very flexible. So for example, what we can do, oops the wrong way we can do. Let's do nine images. Here. We took four images of Jeremy and we decided to create a Jeremy Khan model. This is just with four images. Who knows if this will work? Uh a picture of Jeremy Khan as an astronaut. You want to go to space and then we can do HD map painting. Why not live demo of a new technology who sees knows if it will work or not. We'll find out in a second. The suspense is killing me. OK. OK. How are you doing there? Maybe we should make it professional. Actually, it wasn't professional enough. So it's interesting because it understands all these kind of modifiers. If you do a professional, not professional, you also do negative prompts, you can say not ugly and it becomes not ugly. Um You gotta do not ugly for me. Definitely Jeremy like any of these like, how's this one here? Yeah, not bad. It's fine. Not bad. There we go. Just the four images. We've got Jeremy there and um now we obviously do by banker. There we go, right? So I think, you know, super powerful technology, they can create any image unimaginable just a year ago. That that's not me. That's not you. Another Jeremy's emerged. So mat professional, let's do photo with some 35 millimeter studio lighting to make it good. So, I mean, I think this is the thing, this technology is still nascent, right? This is unimaginable. Just a few months ago, let alone in the last year and it's just evolving at such a rapid rate. Still not, well, it's closer, but this technology is designed to be kind of zero shot where you just put something in, you get out, it's part of a whole design process. So we've seen people take this and create entire movies from it because they take frame by frame like into the Spider Verse and they transform it into an anime movie. We've seen people take this and create architectural designs at scale and you know, again, it's part of the process. It's this amazingly powerful tiny file, we put it out to the world and now people are developing around it and developing an ecosystem that they can then take and build really useful applications because that's what you need actual useful things. This is why the number one App Store app one app Store of the year is Wombo, which uses our technology. Number one on the App Store right now is Lenzo which creates these types of customized profiles based on this version of the technology as well. So I think it's going to be very interesting to see how this develops going forward. Absolutely. And there, I mean, there are skeptics out here sort of would say this is great. It's a lot of pretty pictures. But what's, you know, what's the real business impact of this? So like if you look at movies for example, this will cut down movie generation time by half. So we have an example. Our creative director is shooting a movie in a desert about serial killer with some very famous acts and actresses. Serial killer had a book of all the victims that he had and they were gonna have to hire 25 actresses to fill that book, put them in seventies makeup and you know, fly them out sag rates, make up all this kind of thing. $600,000. He trained a custom model overnight on seventies pictures, it was done. Save $600,000. Wow, that's just one example when you look at video games, movies, other things that transforms in any type of design process, we working with automotive manufacturers to create brand new car designs because you can do these kind of dynamically, you know, you can just type in like an awesome super car in San Francisco and it will generate those but you can customize it with your own assets. This thing when you take the standard model and customize it, it becomes super duper powerful. That's an awesome supercar in San Francisco. Yeah. Wow. And are you, are you thinking of using this beyond just images? Are you thinking about using the same underlying technology for generating other sorts of things? Yeah. So we have the most popular language model in the world G TJ Neo and others from our Luther A I kind of community and we actually do models of every single type. So we have audio models that can generate any type of music. We've got 3D models coming video models. Ultimately, the whole thing is that there's no barrier to creation when you have this next generation of models. Because we move from a big data, we took big data extrapolated for very specific models to these very general models that understand principles like understands the word Supercar, San Francisco, we Twitter about Jeremy, maybe it needs more than four images, but we have bigger Jeremy, right? And this is the next stage in A I and the next stage in real usability of this technology, right? And for you guys, it stability, what's your kind of business model going to be going forward every single creative industry in the world is going to have to react to this because like I said, as of next week, this is 30 images. A second which is basically video brand new experience where you can create anything dynamically. We're going to go to them, take the extra bytes, petabytes of data and convert them into models because again, 100,000 gigabytes to 1.6 gigabytes. Something quite crazy. Yeah, absolutely. Um I said I wasn't going to take questions from the audience, but I think I will because we have a little bit of time who has questions for Ahmad? Anyone have question right down here in front, wait for the mic to get to you, please and please stand and say who you are. What would you say to you? Can you, can you tell us who you are? Sorry? My name is Ian Util Ceo of attention live. What would you say to creatives? Because this can either augment a creatives development of what they do or they can kind of feel like, oh no, this is gonna take my job. So on the optimistic perspective, how are you encouraging creative and creative teams to leverage this technology to do what they do better? So I think this is the thing, what is creativity, right? So this can mash together concepts that so we still have an iterative process to get there. There is just one part of the process. So the ideation phase, the fine tuning phase, the kind of like we have text to image, image, to text in painting so I can remove, I could say it make that car blue and it will make that car blue. In the next version, you have to forget how to integrate this into your processes just as you did with after effects Photoshop and all these other tools because it's just a tool, it doesn't do anything by itself. So what we found is the best artists got even better. And this is what we saw where there was an A I that uh beat Lisa Dollar Go, who was the best Go player? The average Go score. It looks like this over time. Is that because they all got better because they use the technology. This is how you view it as a tool to extend and embrace it because it's here and you should use it right now and then you'll outperform all of your competitors. All other questions over here. Hi there, Ahmad Rishi Gar, a partner at the Mayfield Fund. Um uh You've got a lot of other projects on the website that you're working on. You've obviously started with image generation. But one of the things that's captivated everyone's mind is text to text threading and text generation because of the broad applications. Could you talk a little bit about the future horizon for stability and all the different applications you want to build into and how and how kind of starting with image get you to a place where you can do multimodal applications over time. So that's as a business perspective. When I saw image, I saw it as we going from a 10% to a 90%. Like if I've shown you this a year ago, it's impossible, right? And the fact that it's real time now is also impossible. So you wake up and do some impossible things before breakfast. Text. By contrast, we got 85% to 95% there. And I struggle to see where the edge is going to be when you have a chat bot that can do anything like you use chat GPT somewhere over now. It's fantastic. All the A I models will be like that. Whereas an image I saw that there are giant asset stores of IP that need to be converted because you'll always want Mickey Mouse or Halo or any of these other things. So when you can create everything, something becomes valuable, but it's not enough just to have image because image will be standardized. Instead, what we're doing is we're building the ready player. One oasis hock where anyone can create anything they can imagine. And that's not only media, no more powerpoints, no more slide creation in a few years, just describe it and say make it more peppy, make it nicer and it'll do it automatically recreate Game of Thrones season eight because it was a bit rubbish. You know, these types of things is what will enable seamlessly. And can you talk a little bit about how you build stability itself? Because I think a lot of people were surprised you guys almost came out of nowhere with this fantastic technology. I think there are a lot of businesses out there where they, you know, they talk all the time about the struggle they have hiring A I talent. You've obviously been able to build, you know, really cutting edge stuff. How did you do it? So we built a community when some of the large open source models were open source, the big entities could do it. And I looked at where's the arbitrage ties and the gaps. The gap was super compute data and talent. The talent came from our communities and we hired primo from our communities. So the 1st 96 employees, only one phd so differentiated talent then data, we built large open source data sets like lion 5.6 billion images. The previous launch was 100 million to power this. And then finally super compute worked closely with Amazon to build the 10th fastest public supercomputer in the world. And so that was overwhelming amounts relative to what happened. So 4100 the fastest in the UK was 640. And that unlocked all of this because we allowed people to experiment and play and figure out and make mistakes that then optimized for this crazy model right here. But also other models that were released in our generation to the new language ones, to the others because you need all the components all in one. And that's why creative stability to be like this foundation layer for next generation generative A I and you now have a sort of a partnership with Aws. Is that right? Yeah, so Aws to be a long term partner, now it's accelerating up, we're going to power Generative A I through kind of Amazon and make this available to everyone. It becomes a differentiator for different people's clouds. Obviously, Azure has got their relationship with open A I AWS is you guys, it's kind of an interesting thing. Exactly. And you know, and there, there's different approaches like Azure is quite integrated. Open air is very close to Microsoft. Whereas we're just trying to get this out to the entire world. Great Ahmad, thank you so much for joining us. This has been fantastic, really exciting stuff.