AI meets Fashion: Redefining Fashion In the Digital Age

Space Summary

The Twitter Space AI meets Fashion: Redefining Fashion In the Digital Age hosted by metropolisworld. The AI meets Fashion Twitter space delved into the transformative synergy between artificial intelligence and the fashion industry within the innovative Metropolis ecosystem. AI is reshaping fashion design processes, promoting sustainability, and enhancing personalized consumer experiences. Real-World Assets (RWAs) play a pivotal role in merging AI with fashion, offering unique ownership opportunities. Collaborations between the fashion industry and AI experts lead to innovative solutions for efficiency and trend forecasting. The Metropolis ecosystem stands out for its holistic approach to fashion innovation, blending technology, art, and commerce for a 360° immersive experience.

For more spaces, visit the AI page.

Questions

Q: How does AI impact fashion design processes?
A: AI revolutionizes fashion design by streamlining processes, fostering creativity, and enabling personalized consumer experiences.

Q: What role do Real-World Assets (RWAs) play in the AI-fashion fusion?
A: RWAs are crucial in merging AI with fashion, offering unique ownership opportunities and value propositions.

Q: How can AI contribute to sustainability in the fashion industry?
A: AI promotes sustainability in fashion through digitization, waste reduction initiatives, and smart textile applications.

Q: What are the benefits of AI collaborations for the fashion industry?
A: Collaborations with AI experts lead to innovative solutions for efficiency, trend forecasting, and sustainability in fashion.

Q: How does AI enable personalized consumer experiences in the fashion sector?
A: AI technologies facilitate virtual try-ons, personalized recommendations, and predictive analytics for enhanced consumer experiences in fashion.

Q: Why is the Metropolis ecosystem unique for fashion innovation?
A: The Metropolis ecosystem integrates AI, art, and commerce to create a holistic and immersive 360° fashion experience, setting new standards for digital-age fashion.

Q: In what ways does AI enhance marketing strategies in the fashion industry?
A: AI-powered consumer data analysis enables personalized marketing campaigns, tailored recommendations, and enhanced consumer engagement in fashion.

Q: How does AI promote inclusivity in the fashion sector?
A: AI fosters inclusivity in fashion by understanding and catering to diverse consumer preferences, improving accessibility, and enhancing overall consumer satisfaction.

Q: What advantages do fashion brands gain from integrating AI technologies?
A: Fashion brands benefit from AI integration through increased efficiency, cost reduction, targeted marketing, and competitive advantages in the market.

Q: What makes the synergy between technology, art, and commerce unique in the Metropolis ecosystem?
A: The Metropolis ecosystem blends technology, art, and commerce to offer a comprehensive and immersive fashion experience, redefining industry standards and consumer engagement.

Highlights

Time: 00:15:43
AI in Fashion Design Discover how AI revolutionizes the design process, offering new creative possibilities and personalized experiences.

Time: 00:25:12
Role of Real-World Assets in AI-Fashion Fusion Explore the significance of RWAs in blending AI with fashion, creating unique ownership models and value propositions.

Time: 00:35:58
Sustainability Initiatives Through AI Learn how AI drives sustainability in the fashion industry, promoting eco-friendly practices and innovative solutions.

Time: 00:45:21
Fashion Industry Collaborations with AI Uncover the benefits of collaborations between fashion and AI experts for trend forecasting, efficiency, and sustainability.

Time: 00:55:39
Personalized Consumer Experiences with AI See how AI enhances consumer experiences through virtual try-ons, personalized recommendations, and data-driven insights.

Time: 01:05:17
Metropolis Ecosystem's Holistic Approach Experience the unique fusion of AI, art, and commerce in the Metropolis ecosystem, setting new standards for digital fashion experiences.

Time: 01:15:02
AI-Driven Marketing Strategies Understand how AI empowers fashion brands with personalized marketing campaigns, tailored strategies, and enhanced consumer engagement.

Time: 01:25:47
Inclusivity in Fashion Through AI Explore how AI promotes inclusivity by catering to diverse consumer needs, improving accessibility, and enhancing satisfaction.

Time: 01:35:19
Efficiency and Competitiveness in Fashion Brands Learn how AI integration boosts efficiency, reduces costs, and provides competitive advantages for fashion brands.

Time: 01:45:30
Technological-Artistic Fusion in Metropolis Discover the seamless integration of technology, art, and commerce in the Metropolis ecosystem, redefining the boundaries of digital fashion experiences.

Key Takeaways

• AI is revolutionizing fashion design processes, enhancing creativity, and personalized consumer experiences.
• The Metropolis ecosystem blends commerce, AI, art, and culture to create a 360° immersive fashion experience.
• Real-World Assets (RWAs) play a pivotal role in the fusion of AI with fashion, offering unique value propositions and ownership.
• Fashion industry collaborations with AI experts lead to innovative solutions for sustainability, efficiency, and trend forecasting.
• The digital-age fashion landscape embraces AI technologies for virtual try-ons, predictive analytics, and personalized recommendations.
• Metropolis ecosystem exemplifies the synergy between technology, art, and commerce for a holistic fashion experience.
• Consumer data analysis powered by AI enables personalized marketing strategies and tailored fashion recommendations.
• AI integration in fashion promotes sustainability through digitization, smart textiles, and waste reduction initiatives.
• Artificial intelligence fosters inclusivity in fashion by catering to diverse consumer preferences and enhancing accessibility.
• Fashion brands leveraging AI technologies gain competitive advantages through enhanced efficiency, cost reduction, and targeted marketing.

Behind the Mic

Channel Introduction

Our channel subscribe. All right, we're going to get started here in just a couple minutes. Just waiting for the rest of our panel to come up real quick.

Audio Check

All right, check one, two. Yes, hearing you loud and clear, Brandon. Great to be here. Yeah, likewise. Good to have you here. We'll get started here in just a minute.

Panel Introduction

All right. We're getting pretty close to having a quorum here up on the panel, so we'll get started with an introduction and hopefully we'll get the rest of the people up here before too long. Oh, I see Sergey down there as well. He's going to be joining.

Hedera Ecosystem

So welcome back to another Hedera ecosystem, Xbase, brought to you by the Hbar foundation. I'm Brandon and I'll be your host. Today. We're going to be talking about two of the most revolutionary technologies, DLT and AI, and how they're going to impact each other and maybe some of the use cases that are leveraging them within the Hedera ecosystem.

Panel Participants

So on the panel today, eventually, once we get everybody up here, we're going to have Michael from Tier bottle, Sam from who you just talked to, Sergey from hyper Real. We have Martin from Metaverse, me and Ed from Hashgraph, formerly known as Swirld's labs. So the panelists can jump in here at any time if you have something to add or you can raise your hand and I'll segue to you.

Community Involvement

So we also welcome community questions. So if you have a few questions, just go ahead and request to be a speaker after we get through the first few questions, and I'll try to bring you up.

Panelist Introductions

So to kick this off, I want to have the panelists introduce themselves and tell us about their interest in the intersection in AI and Hedera and maybe a little bit about your business. So we'll start off, since we already had Sam up here. Sam, why don't you kick it off?

Sam from Reality Jam

Hey, thanks, Brennan. It's a real pleasure to be here. And, yeah, honored that you invited me to join today. So reality jam, as we're recently known, formerly Vrjamde, is an immersive technology company, and we are really driving for the integration of web three and AI at the moment through the deployment of our revolutionary new solution to quickly, easily create AI powered virtual humans and then to commoditize those entities, those beings as nfts via our NFT marketplace.

AI Jam Solution

It's a solution we call AI jam. We just turned it on about three or four weeks ago. We've just begun to bring our first couple of client customers on the platform. We're planning a big rollout in the months ahead. Going to be an exciting time for us. And, Sim, you're collaborating with Metaverse me, right?

Past Collaboration

Nothing actually fully live right now. We have connected in the past and had comms in the past, not recently, but very well acquainted with the Metaverse me guys and the project, and definitely a path we should be going down for sure. All right, good to know.

Sergey from Hyperreal

All right, so next up, we're going to go to Sergey from Hyperreal. Can you just give us a quick introduction? Sure, yeah. I'm very happy to be here. Thanks for inviting me. Basically, at hyper real, we empower world class creators to own, copyright and monetize their digital identity.

Digital Humans Creation

So we leverage generative AI to give full ownership of the training sets for generative AI. So we create what we call emotionally realistic digital humans. And we do that through with our experience in the high end visual effects world.

Micro Expressions

We are able to create digital humans that have micro expressions and really express emotional performances. And we also have a way to protect their identity, what we call the digital DNA. So we utilize AI and DLT to basically do this. Understood.

Martin from Metaverse Me

All right, we're going to dig into that a little bit more as we go forward here, but I want to get to Martin and Metaverse me. Sorry, I was trying to put a collaboration together there that might not exist yet, but Martin, can you give us a quick introduction? Yeah, sure. Hi, everyone. I'm Martin from Metaverse me.

Mobile Social Ecosystem

So metaverse me, we like to call ourselves a mobile social ecosystem, I guess, that we're building. So, again, very much like our friends at reality Jam now and Hyperreal, a lot of our stuff is based on generative AI around avatars, but also around digital fashion in a big way, actually.

Ecosystem Development

And that's kind of what we're building. We're building an ecosystem that allows people to create, we like to call them recognizable avatars rather than ultra realistic, and then use those characters to create social content, with a big part of that content coming from generative AI in the form of fashion and animation and some of those kind of things.

Game Development

We also develop games, so we're also developing our own games, having spent far too many years as game developers. So that's what we do. All right, understood.

Consumer Engagement

Now, a lot of these teams are working in the consumer engagement space, and I see that we have Alex down there. He's the head of the consumer engagement fund at the Hbar foundation. If you want to come up and say anything, you're more than welcome. Just request to be a speaker and we'll bring you on up.

Michael from Terabot AI

Next up, I'm going to go to tier bot. And Michael, if you want to add anything and give us a little bit of an introduction, I know we've talked to you a fair amount in the recent weeks, but just a refresh for anybody that hasn't been coming to our spaces the past few weeks.

Introduction by Michael

Thanks, Brandon. Really appreciate you guys having us up here. So those who don't know, this is Michael behind the tier bot AI account. I'm the founder and CEO of Terabyte AI, and I'm also the co creator of Hashnels, the first on graph data permanence protocol on Hedera.

Terabot AI's Focus

And at Terabot AI we're at the forefront of NFT and DeFi analytics, but also ATS based analytics for the Hedera retail community. Well, thanks for coming on today, Michael. I'm going to go to Ed next, but I want the rest of the panel to be thinking about our next question here because I'm curious about how AI and DLT or blockchain is going to or potentially has the ability to enhance each other.

Ed's Introduction

So I'm going to go around. I might grab a few of you right off the bat, but just think about it and if you have any thoughts on that, you can go ahead and raise your hand. But now I'd like to go to Ed real quick. From Hashgraph formerly swirls labs to introduce himself.

Ed from Hashgraph

Thank you, Brandon, and hi everyone. Very happy to be here with this community. I lead the developer relations team for Hedera at Hashgraph and my focus there is to attract and activate developers on the Hedera network.

AI Initiatives

From the hashgraph perspective, we're focused on some AI initiatives for increasing developer productivity. We have a few tools that I'm more than happy to get into later on in the session.

Exploring AI and Blockchain

We're also exploring how to enable developers and use cases that use both of these technologies, that being AI and blockchain. Then I'm very excited about the overlap these two technologies, because there's a lot of upside that these technologies have going forward, and I see them very much as complementary technologies because each one of them has risks and threats associated with them, but they're also very good solutions for those risks that each technology introduces.

Valid Use Cases

And so there are some very valid and compelling use cases that we're exploring with different teams in this overlap. Well, so again, let's dig into that a little bit further. I was just talking actually to the fresh supply company team, and they're using AI to standardize information that goes into their contracts, and they do that over and over again.

Michael's Perspective

It was interesting what they're doing. But, Michael, why do you think there's so many teams that are taking these two technologies and putting them together? Michael from tier bot. Oh, sorry, I missed that.

Question about AI and DLT

I think the audio cut off for a second. What was the question? I'm just curious why you think there's so many teams in our space that are looking to leverage both AI and DLT technology, or Hedera.

Integration of AI and Blockchain

Yeah, I think that's a really great question. For the most part, I've covered this, too, and I've spoken at panels. The idea of AI and blockchain is actually kind of insane if you just think about the premise of AI in general and the costs around it, especially with storage.

Storage Challenges

AI models can be parabytes big, so that's bigger than a terabyte. For those who know, you have your AI model data, you have your training parameters, and you have these constant updates to these models that happened to all the time.

Deployment Feasibility

So the idea of just even deploying an AI model on a blockchain is kind of insane. If you think about bitcoin, for example, where it can cost $1,000 just to put 1 data on a blockchain, it's not even possible or feasible to think about something like that.

Hedera's Cost Advantage

But with Hedera, because of the cost of storage, because of the speed, we can actually reasonably start having conversations about these things, like through the HCS one on graph file protocol that we've built, it would cost you just $200 to put a gigabyte up.

Problem Solving Potential

That's still a lot, but it's less than $1,000 per megabyte. And we can actually have conversations about this and think, hey, maybe this is possible. Maybe we could put an AI model on a DLT, and then maybe once it's deployed, maybe we can start thinking about solving problems like verification and bias.

Trust in AI

One of the biggest things in the AI industry that I've noticed, especially with the different models, is how do we trust that what the AI is saying is accurate? How do we trust that the parameters given to the AI weren't biased? How do we make sure that all those things are accurate?

DLT Solutions

And I think that's really where a DLT kind of comes in, where you can actually, through some mechanism, store your model plus parameters on the DLT itself, and then you can verify the authenticity of the results of that model, and then you can know for sure, okay, this is why the AI responded like this.

Parameters Verification

And these are the training parameters. And now through the power of a DLT or blockchain, we can vote on if we should change those parameters, and then you can tie to economics around them, then you can make it really interesting.

Scalability of Hedera

So a short way of saying is, I actually think that AI and blockchain doesn't really make sense most of the time, but when you're talking about a scalable DLT like Hedera, we might just be on to something.

Concerns for Node Providers

Do you think when you're talking about petabytes and how big these models can be, do you think you're scaring our mirror node providers by putting all this on graph? I think to an extent that is a problem that we're going to have to solve, and I think that we're all aligned on the how we solve that for the Mara note side of things.

Hedera's Advantage

And I think in part it's actually really great that Hedera, unlike a blockchain, actually separates, reads and writes. And that's why our network is so fast and able to submit transactions so quickly, and at the same time we're able to offload that storage concern.

Long-Term Perspective

I think longer term there are things that we can do to decentralize the mirror notes in the ecosystem and provide incentive mechanisms. But I don't think it's a huge negative problem. I think it's more of an opportunity for the network that we can grow around it and make it even better.

Continuation with Ed

That's fair. That's fair. All right, so I want to go to Ed here in a minute just to talk about some of the tools we have. But I do want to continue to go to some of our panelists here and ask them, you know, why they brought these two pieces of technology together.

Sergey on Technology Integration

Sam or Martin, if you have anything to add on this, you can go ahead and raise your hand, but I'm gonna go to Sergey on it. Why did you decide to take the bring these two technologies together?

Sergey's Assurance

Yeah, for us was really important. Our main concern is privacy and authenticity. I mean, we deal, we create realistic, I wouldn't say photorealistic, but they are photorealistic, but they go beyond that.

Targeting Emotional Realism

They morally, mostly emotionally realistic. We create these digital humans for talented celebrities we work with, like Paul McCartney, Madison Beer, Lionel Messi, Pele.

Client Concerns

A lot of our, a lot of our clients, you know, they require this privacy and authenticity, and they're kind of scared when it comes to generative AI solutions, especially deepfakes and the like.

Authentication Solutions

So what we first inform us we wanted a solution that would help us authenticate all of our assets. But then, more importantly, we also wanted to create easy onboarding for our clients on our platform.

Hyperdream Platform

We have developed this platform called Hyperdream with the support of the Hedera foundation, which we're very thankful to, and the easy onboard, the fast and scalable environment, and also environmentally conscious.

Environmental Consciousness

That was also a big concern, and that's something that we found in Hedera. So right now, can we interact with these. Oh, sorry.

Introduction to the Platform

Sorry about that. Can we interact with these, you know, with these different people using your platform? Yeah, the platform is not yet live. It should be in a few months. But what we've been doing for the last four years at hyper real is basically creating these digital humans of this. I mentioned just a few of our clients. There's a lot more and many more that we are working with now that I can still not tell you about. But we have been working for the last four years in creating these assets and creating lifetime value for them. Some of them are, deceased. Some of them, you know, we have the capacity to create, what we call the digital twins, or which are, basically tweens of how they look right now, or we can also go back and make them d age. So we've been creating this, these assets and putting them to perform. I mean, the important thing is that for us, what we want is for these digital humans to perform in concerts, in movies, across the board of all media.

Engagement and Rewards

Also on this platform, hyper dream people will be able to use the platform to create engagement through actions and then get exclusive authentic rewards. And so we've been doing that for a long time, but very soon, everybody's going to be able to do it using our platform. That's amazing. So digital persistence of these artists that can pretty much live on forever. It's really exciting. Sam, go ahead. Yeah, thanks, Brennan. I think there's a really important use case for the integration of web three and AI technology when it comes to b two, b and enterprise. The data sets that underlie these models are ultimately what make or break them. An AI model, particularly if it has relevance to a specific domain of knowledge, is only as good as the data that underpins it. And a lot of the time, those data sets are what is most valuable of all in terms of empowering these entities to come to life, to be amazing, awesome, and to sort of create value.

The Use of Data Sets

And I think that there's a real need for web three solutions to really sort of harness the ip that underpins those datasets where you have a specific company or a brand, for instance, who have specific kind of intellectual property that's contained in the data that's been used to train their particular AI entity. And I think that using web three to commoditize those datasets, and as we're sort of working on doing right now, packaging those up into a singular solution, that's then renders as an NFT, that's something that's really badly needed right now in terms of understanding the value of the data that surrounds the AI use cases. Obviously, some very high profile lawsuits that have sort of come about recently where some of the biggest tech brands in the world have been sued for vast amounts of money by big media companies for just scraping their data off the web and guarding against some of those risks, I think is hugely valuable in terms of understanding how web three can play into the AI technology piece.

Value Transfer and Content Creation

Well, you bring up a good point, Sam, and that's one thing. Even though AI is incredible and can do a lot of things, it doesn't have that value transfer mechanism. And that's another thing that we could bring to it. With crypto and these ability to do really small payments, you could go ahead and have everybody. The issue is the people that are creating the content aren't benefiting from it. Of course, your platforms are trying to address some of that, but if they opt in and then they can get payments using a DLT, that could solve a lot of these problems. So what are your thoughts around that? Yeah, exactly right. You know, and it's then about, you know, sort of commoditization of that data set, you know, okay, so let's say I'm, you know, Brand, who specializes in deep sea geological research, and I have an AI entity that's been trained in a specific domain data set that we've put millions of dollars into procuring and creating.

Commoditization of Data

Well, I can now prove, validate the quality of that data, show that I own it's mine, it doesn't have to be publicly available. It can be encrypted. Then you can commoditize that and make it available to others who also want that same kind of solution to serve their clients customers, to provide value for their particular business. And I think that there's something really powerful there in terms of commoditization and monetization of the underlying data sets that ultimately empower these AI entities to live and breathe. Well said, Sam. Thank you, Martin. Oh, sorry, I lost my voice control. So it's kind of interesting isn't it? We talk about the data on the ledger. And for us, I think we view it slightly differently.

Audit Trails and User Interaction

So the dlts for us, we kind of treat them as, I guess, how would I describe it? So we treat them kind of like audit trails internally. So this audit trail tells us fairly anonymously exactly what a specific user does, how they spend their money, what they visit, what they do, where they go, and everything else. And we can use that data, especially in the gaming space, to create anonymized kind of representations, whether it's for the social graph where, you know, we can connect that user's account to their social activity, whether we can create spending patterns from that, whether we can train models to look at user acquisition, how successful that is, and those kind of things. I think that kind of leads us on to other areas where AI can, you know, kind of interact with these ledgers, things like.

Anonymized Data Applications

We spoke to a company a while ago, actually, that was kind of cool. So what they do is they have these foot traffic heat maps that they create completely anonymized data. But what we can do is we can write that information about, you know, the GPS tracking on a user's mobile, completely anonymous to a ledger, and they can use those kind of heat maps that they can generate using generative AI to work out where might be a good location for a new shop, for example. And these AI models have huge amount of value where the data is completely anonymous. But when we apply it to our own ecosystems and to very specific use cases, there's huge value. There really is. And that applies whether it's to digital twins of people or to anything where we need to interact with an end user.

Digital Identity and Multi-platform Formats

And perhaps the data, as already has been said, might be a little bit personal, private, or somebody wants to keep it under wraps. Great point, Martin. Do you mind just going into a little bit more detail on exactly what metaverse me is doing so we can get a better feeling for your overall vision? Yeah, sure. So I guess very much like reality jam and hyperreal, we're creating, we like to call them digital, not digital twins. We call them recognizable avatars. You know, so we take a selfie, create an avatar, and then we can use that avatar to create social content. And that social content at the moment is to create video and images that you can share to your social media channels.

Avatar Creation Process

So it might be, you know, I take a selfie, I create my avatar, I grab a piece of wearable or, you know, address, maybe from the inventory that we've got in the app, put it on my avatar. And now I can put my avatar into, you know, the real world through ar, and I can dance in front of the Louvre, for example. Or if you're a football fan, I could put on an arsenal t shirt on my avatar, and I could dance in front of the. The old Trafford entrance, which would be kind of highly amusing, create video and post it to my social media channels so we can do that kind of stuff. And what we're using the, I guess, AI and the DLT for is a couple of things. The idea of being able to train a model that can allow us to view what spending patterns are, how users create avatars, how they wear, you know, or what they wear in the app, and track all of those kind of data points to say, okay, this is where a user drops out.

User Behavior Tracking

This is where users do this is where users do that. And it allows us, I guess, on a much more anonymous level, to do what we've been doing for 25 years in games. Figure out why people are playing our games, why they're not playing them, why they're spending, and why they're not spending, which is the ultimate goal of every. Every product we're building, isn't it? It's user retention, user growth, and user spending. So that's kind of the first piece. And then the second piece is we're building our own games where you can actually take those avatars and you can play them immediately in our own ecosystem.

Building Games and Ecosystem Interaction

And again, it allows us to do the same kind of thing across all of those ecosystems and other connected projects and see what user behavior is like for us. That's the big thing. And I think it's interesting you brought up games we're starting to see within the Hedera ecosystem. It takes a long time for these games to be developed and then start to hit the market. Do you think that when you create these avatars, you could see cross game or cross studio use of some of these avatars? Because they are kind of their assets? We already have it. So our avatars can already be played in our own games, which will be releasing later this year, and we will allow them, and we have other partners that are going to be taking those avatars into their own ecosystems.

Collaboration and Interoperability

And this is this. You know, everybody talks about interoperability and cross project support, but we have dozens and dozens of companies making dozens and dozens of. Of avatar solutions. And one thing we found to be. To be really interesting is how much support there is for this stuff. Everybody loves avatars, okay? The end user loves avatars. They want to make an avatar that looks like themselves, and they want to do stuff with it. But actually, there's very little, if any, interoperability at the moment in most cases, and we're trying to change that because I think, you know, if we could take a hyper real avatar, for example, of a famous person, pull that into our app immediately, within seconds, we would be able to have that David Beckham, for example, to use a footballer, we'd be able to have David Beckham dancing outside Arsenal's football ground.

Creative Use Cases with Avatars

How hilarious would that be? Or we could have a digital twin of David Beckham running around in a soccer game, football game for the English, again, knowing that actually isn't the real David Beckham, but it's a legitimate copy of him, if you like. There's some really interesting use cases here. No question about that. Now, I do want to open it up to the community. If you have any questions, you're more than welcome to come on up for it. But, Sergey, I see your hand up. Go ahead, Sergei. Yeah, thanks. I wanted to say about the multiple platforms that I think it's very important.

Multiple Platforms and Consistency

They said here, it's been said that there's many different platforms, and these assets need to be performing in different platforms and different styles. And sometimes photorealistic avatar is what we need. Sometimes it's a more cartoony one, but it has to be a consistency and more importantly, a digital identity associated with it that's been preserved through all these platforms. At hyper real, we do have multi-platform formats, so our avatars can work, perform in pretty much any platform because they're universal assets, and the way that we store them is basically not tied to one specific platform. So we'll be happy to do the Beckham avatar and in your game.

Future Collaboration

Exciting stuff. I'm up for that. Sorry. I've just got to say I'm up for that. We'll talk. We'll talk. Yeah, we'll talk. Yeah. Exciting stuff. That's what we want to see. We want to see collaboration between our panelists and the rest of the builders within our ecosystem. Ed, go ahead. Floor is yours. Sure. I do love all the use cases and all the applications that have been mentioned so far. So I do want to add to this perspective of how the two technologies can enhance each other.

Bi-directional Relationship

Right. I see this very much as a bi directional relationship between the technologies, because we can think about a bucket of applications of AI for blockchain and then all these other applications of blockchain for AI. Right. And I think there's a huge group in or set of applications in both buckets. So when we think about blockchain for AI, I think excellent points have been brought up, right? There's a data integrity. In fact, this intersection has brought data to the forefront. And I think we're going to see not only people care about data integrity and traceability throughout the AI model development and all of that, but also the possibility to create these data marketplaces because data kind of becomes the new oil.

Data Integrity and Marketplaces

And so if you can certify and attest that your data has integrity, you're sitting on a goldmine when you're sitting on data. Elon kind of realized that when he took over Twitter. And so everybody who has some sort of valuable data can attest that it has integrity. And then I think we're going to see these data marketplaces come up and something like the Hedera consensus service is a great candidate to be the seal of integrity for those data marketplaces. I think another one that was briefly mentioned is that content ownership, traceability and compensation for human creators, I don't think there's enough of that.

AI and Blockchain Interaction

If the AI's are generating everything from human source content and those human creators are not being compensated or recognized, I think we're going to see some issues. But blockchain has the ability to solve that from the traceability aspect and the compensation piece by automating value transfers. And one interesting one too, is governance. I think AI can learn from the governance models that have emerged in blockchain and in crypto. I'm talking about Daosh and I'm talking about the drama that unfolded earlier with OpenAI. So when we think about who's developing and driving the development of AI, which is very much a technology that is going to revolutionize in touch many areas, who's driving the development? Right? Is it a few folks? Is it a handful of folks? Or is it something that is community driven? And so I think with Daos and these governance models that have emerged from crypto, we have an opportunity to bring that into AI for more participation and transparency.

Value Transfers and Automation

And then I also love the point about the value transfers, right? AI's are not, or AI agents are not going to have bank accounts. And so AI agents and crypto are both Internet native. And I think that's going to be a great synergy there. When we think about the other bucket of how can we use AI for blockchain, I think that is also a very interesting and compelling set of applications because what that can enable is faster development at both the application level, which I think many people are doing today, but also at the blockchain, at the network and consensus layer level. So I think we're just going to see the game and this competitive market probably pick up in pace. And the other one is more robust development because you can incorporate and automate testing, say, with things like formal methods as you develop the code. Right. You can kind of outsource that to an AI model that can apply those formal methods as you come up with the code. So all of that just to add to the umbrella of use cases and compelling applications that we've been talking about here.

Developer Tools and Enhancements

Fantastic insights, Ed Martin. Yeah, I just wanted to add a kind of interesting point, actually. You mentioned that AI might not necessarily be able to own accounts or wallets. Was that right? Did I, did I miss hear that? No, no, the opposite. So I don't think AI is gonna have a bank account, right? It's gonna have a crypto account because they're both Internet native. And so that's gonna facilitate the value transfers. I agree with you. I think that's a fantastic idea. And maybe at some point we should have a discussion around that's far more interesting. We have to tease it out a little bit. We'll get there eventually. But Ed, I want to go into, for our developers out there, some of the tools that are available to bring these two technologies of AI and DLT together.

Hedera Hivemind Revealed

Sounds good. One of my favorite ones is actually Hedera Hivemind. So if you remember earlier this year, chat GPT, which is accessible to everyone now, they came up with custom GPTs. And these are just, I guess, tweaked versions of the main GPT model that can be tailored or, you know, purpose built or purposed prompt for a specific activity. And so we have this one that is sourcing information from Hedera, documentation from different repos and other Hedera knowledge sources to provide answers to developers. And so that can be used to speed up application development on Hedera. Also maybe to retrieve a limited set of information from our mirror nodes. One of the things that we would love to enable developers to do is to ask a blockchain network in natural language about any entity level information or network level information. Been able to ask an AI agent, hey, how many transactions is there doing right now? Just in plain English? And it understands process the information and gives you answer.

Expanding AI Capabilities

And so that's one of the things that we incorporated at a basic level with Hedera Hivemind, and then we're exploring, you know that there's other initiatives that we're exploring to make that more capable, right? So you can ask harder questions, maybe like, hey, what's the most active NFT collection on Hedera? I think we're going to get to that point very soon in the future. So that is Adairah Hivemind. I put a link in the conversation history that we have here for the space. So that's one of my favorite tools that I use almost every day just for tech support, for questions. As I'm developing demos and applications, I just find myself going back to that all the time. The other one is documentation in our discord server. So we actually have a kappa bot that is also trained on our documentation, GitHub repositories and other knowledge sources.

Kappa Bot's Features

And so whether you're on documentation or reading our docs or on our developer discord, you can use that tool there to ask developer relevant questions, even theoretical questions, if you want to understand more about the network, about the consensus algorithm, about the infrastructure that is also there available to you. And then what I love about Kappa is that if it doesn't know, if it's not sure that it can source the information, it will tell you. I couldn't find this in my knowledge sources, whereas with other models, you have the risk of just getting a totally made up answer with incredible confidence. Those are three tools right now that I've put in the conversation history. The other one that I'll mention, this one is not necessarily Hedera specific, but I found it to be very valuable if you're a developer, is GitHub copilot, it actually turns out to be really good understanding the Hedera SDK and how the SDKs are structured, and it can really help up or help out speeding the development of Hedera enabled or Hedera based applications.

Advice for Developers

I'll leave it there, but there are a number of other tools that we're working on, again to enable people to ask about entity information, historical information, natural language, and then just to speed up that development process if you're actually a developer. So we'll have more information on those as they come out. I'm going to try to get some of the alpha here on what you guys are working on anyway, but we're going to go to Sam here real quick. But before we do, and I want you to think about, you were discussing how we can use these different models and get information out of them. Does it make Hedera a prime candidate for combining these two technologies? Because we have the topics and it's easier potentially to scour the information across the graph. So I'd like you to answer that. But first I'm going to go to Sam.

Hivemind Interaction

Thanks, Brannon. Yeah, this one's for you. I wonder whether or not you've ever considered, you know, putting a face on hive mind. You know, have you ever considered, you know, adding a human dimension to hive mind so that it can and perhaps have a more authentic interaction with an end user, deliver information in something other than a text based format? I think that's a fascinating idea. Right now, with hive mind, we're a little bit constrained, given that it's controlled by chat GPT, you're functioning in the parameters of what Chad GPT enables you to do as they explore new frontiers. We'll probably explore those capabilities too. But I love that idea of making it a little bit more approachable, more human. And maybe that's something that we need to take outside of chat DPT.

Building a More Interactive Experience

So I would love to get maybe some ideas from you. So let's connect maybe after the session and see what we can explore. Well, our solution does exactly that, bolts directly on top of chat. GPT allows you to render a 3d character and then animates that 3d character, gives you a whole range of different voice options, and you can do it all in like ten minutes. So maybe we should chat after the call. That sounds amazing. Let's do it. So I do want to again, offer, if anybody down there in the audience has any questions, you can just request to be a speaker and we'll bring you up.

Value of Hedera Consensus Service

Ed, I want to go back to that. Number one, can you kind of explain Hedera topics and why that might beneficial for these large language models? Sure. So the consensus service is something very unique in the blockchain space, and it's also unique, as far as I know, to Hedera because we're basically exposing that consensus algorithm of Hedera and making it really easy for developers to use. And so if you're a developer is just a decentralized messaging bus where we have a topic and message based architecture so you can publish and subscribe. If you're not a developer, the best way to understand the consensus, the Hedera consensus service is as, say, hashtags on a platform like x, right? So a hashtag is your topic and then every entry that is captured by that hashtag is your message.

Leveraging Information for Insights

Right. And so people can write, people can subscribe to those topics or to those hashtags. And so I think it does position Hedera very uniquely because if you wanted to have information that you want to analyze, say through a large language model, it's very easy to retrieve all that information to maybe store it in your own database and then use that for any insights that a model can give you. You're basically automating that process, making this large scale data available to an LLM too, for any actionable insights. And there's plenty of possibilities you can explore there with that unique set of capabilities that the consensus service gives you. So yeah, I do agree Brendan. I think it is a very unique solution that makes Adair a very good candidate for enabling lots of use cases in this intersection of these two technologies.

Developments in the Ecosystem

Thanks Ed, and well covered. We have Vignesh up here. He is the head of business development in the Middle east and I think South Asia as well. Vignesh, I know you're the point of contact for hyper real. Do you have anything to add to the conversation? Yeah. So you've been working with Hyperreal almost past year or so. So adding on to what Serji was saying, we're really excited to see what's going to come out of hyperdrive and other exciting stuff that's happening is. So hyper has already started working on the hyper models for Mister Yrmanda. So that's happening right now as we speak. So they already started designing the meta band which Mister Raman already announced earlier this year.

Virtual Influence and Legacy

So hyper real are the partners in developing the avatars. So what we're going to do after we are going to develop the avatars, so we're going to have each of these characters have some kind of life. So how are you going to do that? Is we are going to train, we will create custom LLM models where each of these avatars will be powered by AIh, and they can interact among with the fans and engage with the fans. So eventually the idea is to become virtual influencers. This is the idea that we have and this is something that we are right now doing. Mister Rehman has already composed the music and we want to submit for the Grammys later next year. So this is all happening. I know it's taking time, but you know, high quality.

Collaboration in the Ecosystem

So we are taking our time to make it solid and so that eventually benefits the ecosystem and we do something that's really great. This is one of the use case that right now we're working with hyper real along with Shreya Rahman. So this is one of the perfect use case example of how we can collaborate between different companies within the HR ecosystem. Anything that's worth doing is going to take some time, so no problem with that. But, Sergey, I'd like to ask you a little bit about the celebrities you're working with, because I've been talking a lot with my partner, with my company around, you know, how this is all about legacy.

Celebrities and Their Legacy

Is that what they're thinking about, to create these avatars, to enhance their legacy and allow to live on in some way? Right. Yeah, that's. That's one of their main, you know, issues, is create this legacy. And we call the infinite lifetime value that will give them opportunities for not just to monetize, obviously, also to monetize, but also to create an impact. So, for instance, we are working with, let's say, Paul McCartney. He's 80 years old, and obviously he has a lot of fame. He's one of the most recognized artists in the world. But to appeal to a new generation, we've been working with him to basically bring him back as a young singer, kind of the Beatles time.

Engagement with Younger Generations

And that gives him and his audience the chance, or the younger audience the chance to relate more directly with his songs, and not just with what he's doing now, but also go back and listen to some of the songs that maybe they wouldn't initially do. So, yeah, legacy is very important. Protecting their identity online is also very important, especially in the age of deepfakes and all these lawsuits. And there's going to be more that will come. So we only work at Hyperreal. We only work with the IP owners. We do what we call the authenticated AI, which means that we basically generate the optimal training sets for generative AI tools like the future models and others, to generate the best possible output, but also making sure that every pixel of what's generated and also the training data that was generated from is owned by the person that has the IP or their state if they're not alive.

Anticipation for Launch

I can't wait for this. And you said this is just a few months out. Say that again, that you're going to be launching here in a few months, so we're going to be able to actually see some of this come to fruition? Yeah, yeah. I mean, we've been working with a lot of these celebrities. You can check Hyperreal IO, some of the work that we've been doing already on different media, as I was saying before, on tv and music videos and all kinds of other media. But in terms of creating the platform that will allow users to engage and interact with digital humans, the hyper dream platform, it should be launched in a few months. We're still finishing partnerships and content, creating content. But yeah, it's coming. Well, we're certainly looking forward to it.

Exploring AI Integration

Michael, I'd like to go back to you and get back into AI. You were talking about how you're focused on potentially using it all on graph, and I'd like to know the benefits. Why would we do that in the first place? You touched on it a little bit, but I want to dig a little bit deeper. Thanks. Yeah, yeah, I think that's a good segue. I think the huge part of it is the authenticity section and the bias section, right. When it comes to AI's today, a lot of this dispute, I think, is actually fueled by Elon complaining about OpenAI and how closed source it is and how black box it is. And you just don't know where trading parameters are being put into these humongous LLMs. So there could be bias that's political or influence to gauge and turn the public into certain ways.

Concerns Over Bias and Transparency

And these are just vectors that we can't control because we can't see the code, we can't see the parameters. We don't know how these models work. And if we can come up with a sustainable way where we can actually put the entire model on a DLT or blockchain, and then we have a reasonable way, we can then update that model, we can actually verify the authenticity of the results of the execution that these models complete. And I think that's kind of the big one. And coming up with a sustainable approach is also really important because like we mentioned earlier, it's a lot of data, potentially parabytes of data that we would be putting on chain if we're going to execute it, if there's a network that can do it, I think it could be Hedera, just because the scale and storage costs that are involved.

Deploying AI Models on Blockchain

But let's say, for example, that we're actually able to deploy an AI model on top of the Hedera consensus service. We store all that data, it sees data through using standards like HCS one and HCS two, like the topic, registries and files. Now, what we can kind of work towards is creating a token around the execution of that data. Let's just call it AI token. And then let's say we have a decentralized network of nodes kind of operating as an l two, whose sole purpose is actually to utilize standards like HCs one, to pull down the model and pull down the training parameters and execute the AI model in the decentralized version and then the icing on the cake would be to go and verify, yes, with a hash that this was run using XYZ version that's currently deployed on the Hedera consensus service.

Public Trust and AI Control

So now the public can go and say, well, I know for sure that this is in bias, or I know for sure that this is biased because it used these training parameters. And then the community using this token, they can actually vote, okay, let's change this thing about the model. Let's change that thing about the model. And I think for the first time ever, you can have an AI model that's deployed in a way that the community controls, not a black box where we can't control or understand how it works under the hood. I think that's just one of the many use cases with AI and on graph storage that we can pull together if we work at it to provide more transparency.

Advancements in Development Standards

I think the other function too, from the Hashnels perspective and what we're doing in a more immediate sense, can actually be improved and powered through HCS three recursion, which is a new standard that we're working on right now that will enable developers to go and ship fully functioning websites, static websites, games and experiences. When it comes to things like hashingles, it also ships with the capability of executing WASM code. So for the developers in the room, WASM is a way for you to be able to bridge and use languages that you're familiar with. For example, like even solidity, you can convert solidity to assembly bytecode and execute it in the browser. So you can imagine, and this is a bit of a contrived example, where you create the hash that's running in HTML and it's executing a Wassen binary that's actually talking to an AI model.

Creative Potential of AI Integration

And an AI model can have a prompt like hey, what should the color and characteristics of this image be right now? And then it'll pull through the AI model and render in real time inside of your hash pack based on AI generations, so that NfT in your wallet can always be changing based on the prompt of the AI model. I think those are just some interesting ways that you can look at it. I'm sure I'm not the only one, but you're blowing my mind a little bit, Michael, and I'm going to go to you next.

AI and Innovation

Now, Doctor Leemon Baird, the inventor of the hashgraph algorithm, the CTO there at world. He actually has his doctorate in AI. Is he? Now that we're in probably the most exciting time ever for AI, is he getting sucked back into that world, or he's staying pretty laser focused on the DLT space? Lehman, I would say, has enough brain capacity to do a lot right. So he's doing both. He's certainly knee deep and laser focused on blockchain development, on Hedera development, growing the platform and improving the consensus algorithm and the cryptography behind Hedera to take it to the next level.

A Balance Between AI and Blockchain

But at the same time, I had a very brief conversation with him over at consensys and he was telling me about this deep and very dense AI paper that he read for fun. So he's certainly keeping up with all the developments that we're seeing in AI, and he's always thinking about ways in which we can incorporate that at hashgraph for helping developers, for maybe helping the development of our network. And I would see that maybe in the next twelve to 18 months, I wouldn't be surprised if he's gone a lot deeper on the AI front as well.

Engagement and Innovation in Hackathon

So he certainly not lost track or contact with all the knowledge he brings from his PhD in his academic life and early professional life when he was all focused on AI. But he's also sharing that responsibility with other people at Hedera, with other people at hashgraph to let them take those initiatives. So yeah, I would say he's certainly doing a little bit of both, but his focus front and center is Hedera development. Good to hear. I have another question on the other side of the spectrum. We just had a hackathon, or we just wrapped up with the submissions for the hackathon.

Highlights from the Hackathon

Was there any AI teams products that came out of that you're interested in? Fantastic question and great topic. I'm super passionate about this topic. So we just had this hackathon from July 22 up until yesterday night. Lots of submissions came in. We had thousands of developers join for that hackathon. In fact, one of the tracks was AI development and blockchain. And so yes, lots of projects were submitted. You know, I'm still going through all the different submissions that we received. It's probably going to take some time to vet them all.

Exciting Project Submissions

But yeah, I actually saw, I'm not going to share like the team name or anything, but there was this very compelling solution of using the generative AI to speed up project management task definition and then using the Hedera consensus service to track the development and delivery of that project, and then using their hbar payments to automate the value transfer once that project was completed to the satisfaction of the project owner and the developer delivering that project. So I thought that was a very interesting solution. There were a few others that I also saw that come to mind, but, you know, I'll probably share more details in a detailed blog once we've had the opportunity to pass these projects to the judges and announce the winners, but certainly lots of compelling projects in that track.

Closing Thoughts and Community Engagement

Yeah, I'm certainly looking forward to everything that comes out from that. So we're getting towards the top of the hour. I want to give the panel an opportunity to give some final thoughts. Sam, we'll start with you. Is there anything else you'd like to pass on? Yeah, I guess I just like to kind of express the, you know, my awe, I guess, at the vibrancy of development and the vision that has been espoused by so many people on the space today. You know, there's been so many incredible ideas that have been expressed here, and it's fantastic to be building inside such a diverse and vibrant ecosystem with so many incredible builders in it.

Invitation for Collaboration

And I definitely encourage anybody that's on the call today who's interested in the AI space to get in touch. We're very keen to collaborate with anybody else in the space. We've built a solution that's very easy to deploy, very easy to plug into things like hive mind. I don't know that much about hive mind, but using that as an example, I'm pretty confident that we could have hive mind having live conversations with people as a semi human entity in a day or two. Yeah, metaverse me, guys, obviously we've spoken a lot in the past, not recently, but I'd love to reconnect and see how we can refire up our previous collaboration.

Building Community Cohesion

As Brenda mentioned at the start of the call. Yeah, very keen to join more dots between the community here and get a bit more cohesion around AI development in the space. Well, we're looking forward to what you come out with. Sam Martin. I'm going to give you the floor next. Is there anything else you'd like to pass on? Yeah, I'd just like to say I think, you know, I agree with Sam to a large degree.

Potential of Hedera Ecosystem

There's a, in fact, completely, rather than a large degree, there's a huge amount of. I think the Hedera ecosystem is interesting. There seems to be a lot of innovation in the ecosystem. There's a lot of projects, I think, that could potentially collaborate. And I think the potential there is to take a kind of a, what I would feel at the moment is a chain focused on kind of defi and those kind of projects to move into some very exciting other spaces. You know, the gaming space, which obviously is one of our big interests, is a huge open area, I think, for Hedera.

Future Opportunities with AI

And I think the application of AI on the DLT is going to help with that massively, especially around analytics for user acquisition and that kind of stuff. And then I think there's the potential there to kind of create new projects, entirely new projects from some of the stuff that's being discussed. I love a couple of the comments around some of the, especially the hive mind stuff where you can take personality and plug that into something. Those kind of uses, I think are going to be incredible.

Collaborative Potential with Celebrities

And then the ability to track ownership of data across all of this, especially when you've got celebrities involved. I think there's huge potential here, and I think the collaborative ecosystem that Hedera is building, you know, especially around consumer engagement, which I'll poke Alex for, I think there's potential there. I think there's huge potential here to do stuff that's not being done on other chains.

Final Thoughts from the Panel

Well said. Well covered. All right, Michael, I'm going to go to you next. Is there anything else you'd like to pass on to the community? Thanks, Brandon. Yeah, I just want to say that I feel like I learned a lot in this call, and it's great to see so many companies building in Hedera ecosystem and approaching it from the AI angle. I think it's hard to see from the outside in sometimes, but we have so many incredible builders fostering this ecosystem, and I'm just really excited to see where things go over the next few months to the year, coming into some of these use cases going live.

Opportunities in Technology

Well, thanks for coming on today, Michael. Ed, I'm going to go to you next. Is there anything else you have? I would say that we're early in both technologies, and that means there is a lot of opportunity. So one of the ideas that I love from the thought leaders in the space that I listen to is that AI today is the worst that it's ever going to be. Right? It's only going to get more capable, more adopted from here. And so I see that as very exciting and very positive.

Invitation for Collaboration

The other thought is just get in touch. You know, if you have questions on use cases and this overlap of technologies, or if you have questions on how to use Hedera to build in this overlap in this intersection, myself and my team were more than happy to help and have those conversations to either guide you, help you use Hedera, or identify opportunities for collaboration. And then the last thing that I would say is try out the tools.

Exploring Tools and Hackathons

We mentioned Hedera Hivemind today. The documentation has an AI component, same with our discord server. And then we're also going to have an upcoming hackathon around November. So even though the one from the last month just closed, there's going to be future opportunity to get hands on, right, to try out tools to build things. And it's pretty low risk when you build on a hackathon. And there's only upside because you can always win some compelling prices.

Inviting Experimentation and Innovation

So take those opportunities to experiment, to connect, and to build on Hedera. Thanks, Ed and Sergey, I'm going to give you the last word. Oh, thank you. Very briefly. Yeah, we live in very exciting times where we, the organic humans, are starting to collaborate with digital humans and with other AI agents very soon. And having an ecosystem like Hedera, with all the innovation that is on the DLT, it's something amazing for us to create tools on top of this great technology.

Looking Forward to Future Collaborations

And, yeah, we're looking forward to collaborating with, collaborating more with the HR foundation, with Hedera, and with anybody in this call, any partners that want to work with us to define the future of digital humans and how we, organic humans and digital humans, are going to be interacting and monetizing in the present and future. Well, it's definitely an exciting future. I want to say. Sergey, thank you so much for coming on today. The rest of the panel, we really appreciate you guys giving us your time.

Appreciation for Engagement

Certainly appreciate all the people who came on and listened in.