Evo Heyning and Jon Radoff discuss artificial intelligence, virtual beings, creator economy & DAOs

Evo Heyning: The metaverse is what we created to be.

Evo Heyning: It can be the sum total of all of our dreams and all that we want to see in the future. It can be the Protopian idealized space we want to create in reality. And it can be the place where we test out those ideas. So whether you're interested in games or work, digital twins, enterprise collaboration, just creative play and artistry, music, theater, all of it, finds a place there. And right now is a great time to experiment with how you can create value with others. Get your own ideas out and maybe do some really interesting experiments, do a little R&D on your own and test yourself. But also think about how you can connect with others, collaborate and maybe make something that is bigger than all of us.

Evo Heyning: Evening sits down with Jon Radoff

Evo Heyning: in this fireside chat about the metaverse. Evening is a creator, interactive showrunner and producer, CEO at playable energy. A brilliant thought leader on many aspects of the metaverse, but particularly around AI virtual beings and avatars. Let's jump into this fireside chat. We are getting ready for another fireside chat about the metaverse with Jon Radoff and Evo Haning.

Evo Heyning: I'm going to let you guys take it away.

Jon Radoff: Hey, Evo, so glad you could be here.

Evo Heyning: Hi, John. Great to be here. This has been a journey and it's great to see you today.

Jon Radoff: Yeah, and just to give a little bit of background here for people who are tuning into this, I met Evo in the virtual beings group over on Facebook. And she posted this pretty amazing message about the evolution of artificial intelligence, virtual beings, machine intelligence. I'm not going to read the whole excerpt here, but you compare it to mushrooms, crystals, kittens. I just loved the metaphors. And I was like, I have to get her on a video and talk about this stuff. Because to me, the metaverse is this convergence of a lot of trends, it's creativity, it's artificial intelligence, it's the immersiveness, it's all those things coming together. I mean, maybe we can start there. How do you conceive of the metaverse, this term that everyone's throwing around?

Evo Heyning: Mm-hmm.

Evo Heyning: So I like to picture the entire digital universe. And that includes many different manifestations and ways of exploring. So we have that VR metaverse that we've seen in the sort of Snow Crash Ready Player One kind of storylines. We have the layers of the metaverse we understand through augmented reality, right? So a snap layer on top of our photos. These are our layers of metaverse experience, the same way something like a spatial audio on WebXR. That can also be these sort of nested, immersive and interactive layers that form something that is not just a 2D web. Obviously, we're moving into not just a 3D web either, but a multi-sensory immersive web. So what do you add all of that up into? What does that create together? Well, that's the metaverse.

Evo Heyning: That's this universe of potential, basically.

Evo Heyning: What creative potential we hold together.

Jon Radoff: I love that you included spatial audio in that description, because I've told people on the social audio platforms like Clubhouse when I've jumped into some of those chats. I've been like, this is the metaverse too. This is part of it. I'm curious how you think of that.

Evo Heyning: Well, so within my work at Open Metaverse Centraloperability, we are working on those layers of the experience. And we started the bottom, and we start with ethos and governance, and how do communities actually connect at a fundamental level? And then we start talking about infrastructure and communications layers and creativity layers, and how all of those things come together to form stacks. We understand what the WebStack has looked like up to this point. Now we're talking about a multi-dimensional WebStack. And audio plays a role in that. And obviously, things like how we share assets, the transactional layer. So that's where crypto and NFTs fit into this entire multi-dimensional approach to exploring the digital landscape.

Jon Radoff: Awesome. So I want to talk about mushrooms and kittens and light and nature as a description for intelligence. Because I think a lot of people think about machine intelligence. They anthropomorphize it on the one extreme, or it's an algorithm to show us advertisements on Facebook at the other extreme. Everybody here understand the different ways that we might think about intelligence and the role that that will play in the Metaverse.

Evo Heyning: Sure.

Evo Heyning: So I'm a creator.

Evo Heyning: I'm a storyteller. And almost 20 years ago, I started creating a world. And that world takes place in a near future where our intelligent ecosystems are a part of our everyday lives. And we live with them. We live within them. And so I began thinking about, well, what makes up a computer today? It is mineral, right? It is not necessarily animal. It's mineral. And all of those, let's say, if you look at a quantum computer, the gold and all of the different elements that come together, how specific properties they act in specific ways. So if you take that and make it bigger and begin to think about, well, what could a bio-computer be like? And how would that be different from the types of computers we work with today? Because right now, we understand computers to be very much the sort of boxy, most of our devices are these boxes, or the flat screens and these sorts of things. But they are not necessarily thought of as organic. We're moving toward a much more organic understanding of computing and of intelligence. And that's intuitive. And that's something that we participate in because we're already a part of that natural world that's intelligent all around us. And so I began thinking about, well, OK, we have computing networks. And we have my Celial network. And we have metrics of crystals. And each of those have fundamental properties that are perhaps in cooperation, like a symbiosis. So what would biomimicry apply to the metaverse look like? How could that grow entire new worlds? And what would that look like in a term of play? Could we bring people into those environments and maybe generate and evolve them together? So I began creating that framework and a series of stories called Manor Mata back in 2005. And so I went into second life to build these giant crystalline intelligent structures, thinking that I could maybe, you know, it was the early days of AI. And I wasn't able to really bring the sort of intelligent objects to life that I had wanted to see. Because I could see that AI doesn't have to be an individual, like a persona or an avatar.

Evo Heyning: An intelligent ecosystem is also a form of AI.

Evo Heyning: And there are lots of ways in which we may be living within AI in the future. And so our relationships to these things are really important. The way we relate to an AI right now in terms of sort of algorithmic bias and all of these sort of regulatory conversations we're having in the public comments right now, those are going to set the framework for how we work with the natural world, the natural computing world, right? Because just as you and I are having a conversation, we're going to be conversing it with a wide number of AI in the future. We already are. And so as we begin to take a look at that broader ecosystem and the role we want to play in it, right? We are cooperative beings. We like to work together and we're interdependent beings, too.

Jon Radoff: Yeah, I love the idea that we can expand our understanding and explanation of just intelligence. So when you mentioned fungi and mushrooms and stuff as a form of intelligence, a friend of mine once said, hey, Bitcoin is actually a form of intelligence. And to be clear, I'm not sure I totally buy it. But his case was this. It was like when there were first flowering plants back in the Cretaceous period and plants kind of formed this alliance between insects and mammals and spreading their seeds around, that's kind of like Bitcoin. To him, Bitcoin and other people have created this metaphor like Bitcoin's the mycelium of money. And Bitcoin is actually the intelligent thing

Evo Heyning: and we're out there kind of serving it

Jon Radoff: by adding energy to the system.

Evo Heyning: So I think it's probably not quite right.

Jon Radoff: In my opinion, I'm curious about your thoughts on it. But what I like is the metaphors in terms of maybe how it helps us think about the problems, think about the connections between computers and machines and just different ways to challenge ourselves about how to build out a lot of these systems.

Evo Heyning: I think what you're from is also pointing to is that the collective has some sort of consciousness

Evo Heyning: of it so.

Evo Heyning: And it may be present about the individual Bitcoin as an intelligent actor in an ecosystem as much as the intelligence

Evo Heyning: of the collective is evolving toward very specific things.

Evo Heyning: And it's testing ideas around decentralization, obviously. But it's also testing ideas about sharing and what we value together.

Jon Radoff: So you were talking about decentralization.

Evo Heyning: Yes.

Jon Radoff: You are also part of an effort to try to make the metaverse a lot more interoperable. You talked about some of the structures we need to enable governance, maybe asset exchange. So to you, what is the role of decentralization and what we're calling the metaverse?

Evo Heyning: Well, I think we got a very interesting life lesson in why decentralization matters when Facebook, WhatsApp, and Instagram all went down at the same time. Those who had all of their social graph and their contacts and their messaging tied up in those ecosystems all of a sudden could not reach their people. They couldn't make their plan. They couldn't do the thing they meant to do. Their businesses failed for that time. And so we see we're having a single point of failure for all sorts of resiliency challenges throughout the comments, not just for the stakeholders, but for everyone else around them too. And so decentralization, to think of it as a broader approach

Evo Heyning: to resiliency, if you're thinking

Evo Heyning: about millions of creators all sharing their stories in networks like YouTube or Twitch event, right? So story driven networks where the metaverse stories are coming into the public. Now you can imagine those stories are going to take on all sorts of wild ideations. And some of them may not be appropriate for all types of audiences, right? If you have just one central pipeline for media, then it's automatically going to cut out everything. Everything that's not appropriate for this audience, that audience, that audience, that audience, right? What's happening is that we're seeing that niche communities each need their own types of media, right? So creator-centered communities have formed that are appropriate for these niche communities, right? And this has happened in queer communities for a very long time because safety matters, privacy matters, right? This is the difference between life and death. If you're a stateless person, if you're a refugee, if you're marginalized, you've already had to do this sort of hacking around the problem space, right? You understand why it's a problem.

Jon Radoff: So basically all of the products of Facebook went down a couple of days ago. You mentioned three of them Oculus. There was another one. A lot of people think of Oculus as maybe Zuckerberg's vision for the betterverse in terms of just being able to enter a virtual reality version of it. And it's funny you say that because I posted this exact same thing on Twitter, also a centralized platform, by the way, but I posted that this really reminds us the danger of centralized systems when you build lots of systemic dependencies on it. So yeah, I mean, it was an inconvenience to some people just to not be able to go and look at your friends' photos or whatever it is that you normally just log in for. But I think it's important to also realize that there's a massive business ecosystem around it. It's not just placing ads. There's a lot of applications. All of the online applications, for example, of Oculus all depend on being able to get into Oculus and be able to do a Facebook login. So there's a lot of businesses that are just completely dependent on Facebook being in law operates. So that too.

Evo Heyning: I mean, there's three losses that need to now be managed, right? How, if you've got that single point of failure, that's a huge problem. I thought the joke very funny was like, how do you, do people stay trapped in the metaverse when it went down? Like is it like a snow crash level situation? Like if you were in VR and that one, you'd be dying and you'd get in the back. All of those funny ideas about, but it's quite serious. And I think you pointed to this in previous articles around portable communities and a connected social graph. So the metaverse I want tells me where I can go find my friends all across these different platforms. And I can get some of that right now in steam, but I can't get to a lot of that information. And it's certainly not enough to develop any sort of interoperable strategy.

Evo Heyning: So we have to use Discord as our communications layer

Evo Heyning: so that we can actually stay connected and have that portability that we're craving.

Jon Radoff: So let's go back a little bit because you mentioned second life. What second life is trying to do a thing right now, a little bit they're like, hey, don't forget about us. We were at the metaverse too.

Evo Heyning: Yeah.

Jon Radoff: And that's fine. I include them in my definition. Actually, my definition is much more about trends and it's about the future of the internet, not you get to be included in the metaverse club or not based on when you were made. So second life can be in the metaverse, but what do you think the differences between second life of then and maybe where it's going other than maybe the obvious, which is that's another centralized platform. But where are we going from here? And you were a content creator there. So how is that going to change?

Evo Heyning: Yeah. So to be clear, I was a content creator there 10 years ago.

Jon Radoff: And that's when everyone was a content creator.

Evo Heyning: And I did second life. I stay in touch with a number of the communities that I helped begin. One of them, the nonprofit commons in second life is still active and they still meet and they've been meeting for 15 years. And so there are certainly OG metaverse communities that are still robust and active who are doing the R&D, the experimentation, the university layer work on accessibility on, again, that niche community service, which is very important because honestly, what keeps people coming back are those communities, right? Generally, we might have a one in done metaverse experience like a concert or a big show a festival. But why we keep coming back is generally because of the connected community that we find. So the communities that really understand that and second life was the earliest in terms of a 3D community that understood the importance and value of the social graph and staying connected to it. And it gave users control to make their own objects and sell them, right? Which every other, you know, roadblocks has taken off and really taken that model and made it available for kids. So they've made it sort of sanitized and safer for kids. But they were the first ones to really sort of prove the model.

Jon Radoff: Yeah. And all these centralized platforms, I don't want to talk about them as if they were bad or wrong or anything. I think that decentralizing all these features is actually super, super hard. Like that just takes a big investment, a lot of coordination, a lot of pieces that work really well. The original internet, you know, was built on fairly simple by today's standards anyway, building blocks like the domain name system is decentralized and then came along the World Wide Web, which is inherently decentralized for the most part. But then stuff got harder, right? So then you had people trying to build e-commerce websites. So you eventually end up with like a Shopify to make that super easy. So you don't have to build Amazon.com by yourself or web authoring systems. For example, like, you know, now we've got square space and wits and things like that.

Evo Heyning: I think creative tools to layer in and to begin to, again, build that sort of multi-dimensional world from these 2D concepts into 3D and 4D.

Jon Radoff: Yeah. But a lot of the common property of a lot of those things is that their focus was enabling creators and also trying to just figure out what their economics would be around that so that they could get paid for creating that environment. And to end up with like either call them Wild Gardens or company towns around content instead of the decentralized web. So in the more 3D immersive space, you've got, yeah, second life, now we've got Roblox. You've got a lot of things like that. So it kind of repeated itself again and I'm wondering what are the pieces going to be that we need to actually enable people to kind of control their own destiny, own their own creations on their own worlds, even within an internet that's going to get more and more real time, more immersive.

Evo Heyning: So there's a handful of fundamentals that need to change. My colleagues at OpenMetvers and Veraprability have been working on changes and really updating the GLTF spec. And there's some interesting work happening at that layer. So for those who aren't familiar with GLTF, it's basically like the 3D JPEGs. If you want to take a picture of that in your mind, these multi-dimensional assets are what we share between, let's say, NFT markets or between Metaverse platforms. Now, if you're an artist of 3D assets, you might upload a sketch fab so that you can resell your work across multiple ecosystems, or you might do an NFT and Mint and do some sort of limited edition release, however you choose to generate value or attention on that. But they're not portable. Almost none of these assets are print once and done. You're always iterating. So let's say I publish a piece of art to a virtual world. I might still have to go back and tweak it many different ways in order to recover it to a different virtual world. The artists who did the 3D Metaverse is, for example, spent hours just working on the shaders and the interactivity because that portability of assets isn't easy to do yet. So there are absolutely things around using WebXR and potentially aspects of the NFT market to deal with that problem. And there are great experiments happening in that space. We're going to do some demos on those within OMI later on this fall. So specifically, how do we use the Open Web to share those assets? We're starting to see, for example, NFT galleries where you can then go buy the asset you want and then bring it into the world you want. And you have permission to do that. That's an early stage. That's the individual. And does that roll up into with collectives and with paired programming and generative art tools? Well, that's updates to metadata and how is fractional ownership tracked? So there's a number of different experiments in infractional ownership, basically. And then there's the other side of it, the publishing side of it. And WebXR has not a great on-ramp right now. So how we publish to the Open Web is going to fundamentally change in the next year. We've seen some great examples of what that can look like. And I think we're going to continue to see, like the NFT market has been a generally flat 2D market up until this point, except for the somnians, the decentralized lands. We're starting to see much more of a 3D and 4D approach to these markets.

Jon Radoff: And a lot of people listening to this or watching this may not even be familiar with WebXR. So we should just take a moment on it. Would you care to just... What's the 4D on WebXR?

Evo Heyning: And I feel like I'm not an engineer and I'm not the best author to do so. But as a creator, when you are looking for the Open Web, you have sort of limited options at this stage. If you're a VR creator, for example, you're going to be generally working within a game engine, like the work you're doing, John, working with the Unity game engine. However, it is often like you're stuck within specific publishing ecosystems. By publishing using WebXR straight to the Open Web, you can overcome some of those sort of silo boundaries and make your work more accessible. So in education, for example, we published a piece a couple of years ago, it's at wildhifengacal.org. And that's a tour to the Amazon to see where your template comes from. It's an educational piece, it's 360, it's an interactive piece, that's designed for kids. And it didn't make sense in one particular marketplace, it wasn't money driven. It was funded by a public third party group in order to do an educational piece of work. And so we don't even have great distribution networks for these things yet. I think we are going to see... When we talked about DAO's and decentralized work, decentralized media networks and decentralized metaverse media is where we're in terms of how do we own our own media together and create value there.

Jon Radoff: Yeah, and I want to talk about governance and DAO's and all that stuff. Let's go there next. So I've had a couple of conversations in this series with thought leaders around the metaverse already. And it's funny, DAO's and governance has come up the two other times as well. And I suspect a lot of people are just really curious about it. So just high level, a DAO stands for decentralized autonomous organization. Now, when I've spoken to Amy Jo Kim and Rath Koster, they had kind of a similar set of opinions on it, which is number one that governance isn't a technical problem. I think they both agree to. It's more of a social problem. But also that they were skeptical of these kind of systems being adequate or helpful for managing like things at the complexity level of like an online game world or something like that. And I'm not so sure because I just... Here's my own opinion on it. And I want to really hear yours, which is, first of all, having a DAO doesn't immediately mean turning over total authority to everyone who just happens to have the DAO token, which gives you voting rights, you can establish what you think the appropriate governance is, whether that's the ability to vote on proposals, whether it's only certain kinds of proposals, whether it's electing a council of elders that help advise, but it's not like you start to hold these tokens and you get to have a say in every single thing that goes on at the level of say, a designer working on something, for example. So I feel like we just need to really explore the space in terms of what DAO's can do, how you structure things, and it's very experimental right now because actually people haven't been doing this for very long. But I don't have the negative reaction that it's somehow opening up Pandora's box of every one's gonna work like game designers and world building in these settings. So, and what are your thoughts on this? Because you've spent a lot of time thinking about all these things.

Evo Heyning: I think we will need a hybrid approach for the foreseeable future. Obviously there are things about the web three and DAO movements that are essential

Evo Heyning: and are really important to deal with breakdowns

Evo Heyning: in governments and breakdowns between sectors as well. Now that includes making sure that one company doesn't have full dominance over the field. And so some of the issues around antitrust can even be approached through a DAO based approach. Now, just to step back, a DAO, a decentralized autonomous organization can be sort of held by many, many people. And those stakeholders are given different levels of participation. And each group gets to decide how much that participation matters, right? What are the decisions the group gets to make together? And you're right. For now, we haven't seen as many fully decentralized DAOs. We're seeing this sort of give the community some autonomy and then perhaps put other aspects of the autonomy within another sort of independently held and that can be an AI, that can be any number of things. And so these sort of paired programming approaches where you see the community of stakeholders having a role, the advisory or the people who actually created a thing, having a piece of that, that's it, too. I'm very excited for how this can change ownership because I think we gave away too much of our creative ownership, especially if you look at the last sort of 10, 15 years of social media, we gave away most of our value, right? We just sort of openly just sort of shared all of our videos and our stories and our assets and our art. And we just sort of did it. Like it was fun for a while.

Jon Radoff: Like we're giving half of the Euro revenue that this will produce to you too.

Evo Heyning: And then we're going to be in the middle of the list. Oh wow, they capitalized on our data and maybe in ways that don't feel good to us, right? So reclaiming our own value includes things like self-sauvered identity. And this is one of the things we work on within Open Metaverse Central Ability is, how do we implement self-sauvered identity across Metaverse platform so that I can maintain control

Evo Heyning: over my own data, my own environment

Evo Heyning: and my own creative work? Because right now, if I go into certain platforms, I give it all away. I don't have a choice. And Dows do provide some way to mitigate that, but only if they're designed to do so. And so whether we see autonomous organizations that respect the quality and the value of the individual creator as well as the whole sort of market, that remains to be seen. I don't know that that's what's going to happen yet. And I think it's important to keep asking the tough questions about does this engagement add up to the thing we want? Or does it just create another sort of money-driven approach that forgets about the humans in the process?

Jon Radoff: Well, that's a question. What happens when you blend together the idea of creators collaborating and whatnot? But also people who want to just maybe not be the creators, which is participate in the world and live in various worlds, and then create essentially a economic incentive around it. Because Dow tokens usually have a value, and they're traded, and some people are just speculating on the Dow token. They're not just there for the governance rights. So how do you think that dynamic will play out or without predicting the future necessarily? What are the risks or issues with that that you see?

Evo Heyning: I can give you one example from my story world.

Evo Heyning: So in that world, you can have different levels of participation.

Evo Heyning: You can be a full character folder, and have a full character that is embodied that you can come in and participate with and on. You can be more of an audience member. So participation doesn't have to be like in the middle of the story. Participation can be like, I'm a part of the walls. I'm a part of the environment. And maybe I make certain things happen. So I'm like the guy behind the walls at Blue Man Group who's talking to you through the tubes, but you don't know. Right? You play with those sort of layers of interactivity and permission and give people different permissions to play. So we did this in an AR game that I designed, but we didn't put it forward after play testing. We did a project around intergenerational play using sort of a nianctic Pokemon Go-like dynamic using seeds and gardening and permaculture as the goal and the reason to collaborate. In that game, it was called Seats the Game. This was seven years ago now, early AR. We had grandparents, mastergarders, and teachers creating the challenges for the young players. That intergenerational play where they are actually being challenged by other people in their community up to the stakes. And it made it far more fun so that they were all being incentivized, even though the incentives were obviously uneven. So you can really play with intergenerational play through these dynamics. Layers of participation where everyone holds one piece of a collaborative stake in the puzzle.

Jon Radoff: All right. So you re-open to the door to characters, a little bit more around playing the role of characters. AI has always played a role within computer games. And one would say the game is sort of one of the dominant themes

Evo Heyning: of the metaverse right now.

Jon Radoff: So we've got this idea of non-player characters driven by AI. So we've kind of had the protoversion of what some people are starting called virtual beings.

Evo Heyning: I was hoping you could maybe shed some light on that.

Jon Radoff: What's a virtual being? And where are we going with this? What's cool about it?

Evo Heyning: Hey, John, did you see the movie Free Guy?

Jon Radoff: I haven't seen it yet. I want to see it. But I've seen the trailer. So I know the basic premise.

Evo Heyning: An excellent example of a virtual being. It is a story of a non-player character who

Evo Heyning: comes to life in some form.

Evo Heyning: And that evolution of intelligence from non-player character into a virtual being that people form an attachment and a love for, right? That has some real autonomy and ability to publish and ability to create. So we're seeing this again. This evolution from just the avatar as something that I embody and then act through, sort of puppeteering in this sort of second life movie. Into now we have obviously lots of V tubers and virtual beings who are maybe not a human inside. Maybe it's all AI. Maybe it's a hybrid. Sometimes these lines are very blurry and unclear, right? Obviously, lots of people are playing with unreal and meta humans and doing things that sort of scan from the physical and put it onto the digital. We're going to see lots and lots of these types of sort of other worldly types of relationships with these types of characters because they feel more real to us now. We begin to see each other in them. And so the relationship with them is going to change. I'm excited to see if this changes our narrative around things that are alien or foreign. Because if we start to see ourselves more effectively and as alien or foreign, does that change the interpersonal dynamic? Are we more likely to feel connected or even friendly toward it?

Jon Radoff: Very interesting. So there's a couple of threads forward with that. One is just the AI, the machine intelligence that might be the mind or heart of a virtual being. There's also the human controlled aspect of it. I'm thinking of like, you know, Kodniko has made a huge movement around this. In fact, Josh and I were just talking earlier today. He was like, hey, we should do something in Roblox or a 3D environment. And I was like, sure, I'll get a Rococo suit. And like, I'll control a virtual being. We can go all the way in. The thought that just occurred to me, though, is it's almost like what you described the end point where the intelligence on the inside may not be a human. It could be a lot of things, I suppose.

Evo Heyning: It could be a hybrid of all of those things. And we may not be able to decipher what's on the other end of the future. So all of these conversations and governance around black boxes and algorithms that are happening in public this week will also be applied to this space, done the line.

Jon Radoff: Yeah. And I'm just thinking now, the more and more Kodniko's there are out there that are playing these characters and creating these experiences for people. So Kodniko will issue some B-roll here, just so everybody sees what we're talking about. But Kodniko plays a character essentially on Twitch and has people come into her channel and responds to things. And you can push buttons and make her do all kinds of crazy stuff. And she interviews people. It's super fun. Just check it out. But now I'm thinking as there's more and more Kodniko's in the world, people are going to start getting accustomed to relating to this virtual representation of a person or a being. In the same way that I feel like we have gotten more and more comfortable over time with virtual property, whether that was virtual goods in an online game or the things that you were doing in second life or all these things that are happening in Roblox or even cryptocurrencies, crypto assets, NFTs, all that stuff. We have become very accustomed, or at least the last couple of generations of people on the internet have gotten super accustomed to that. Now, the next frontier is maybe virtual beings or something that they also just relate to. And it will, yeah, will it become less important whether it's even a human?

Evo Heyning: I think we are going to evolve to that play. So shows like alter ego, we're going to become more and more about what the person shows to become and why. The hybridization is very interesting to people. Some people really like the choices that are happening behind the scenes. But even in fiction and story worlds, I mentioned the writing I was doing earlier. If you can imagine nine human characters, multiple different types of AI embodied characters, animal characters, and other natural elements all in conversation together. Now, I think we're going to see more and more of these types of storylines where it isn't just real life and it isn't just virtual. It is the conversation between them and the dance between them that starts to unlock new things. And I think that's collaboration, especially. It's also pattern recognition, it's decision making and being more effective at things like disaster response and doing things together that are really hard.

Jon Radoff: And as a storyteller myself and someone who's made games,

Evo Heyning: I'm just imagining the possibility space of,

Jon Radoff: I start with the world and I create a world and I bring these beings into it. And I sort of unleash them and see what they're going to do. And it's almost a co-creation process with these virtual beings as opposed to me having to sort of do at the detail level of absolutely everything.

Evo Heyning: Right, right. You're describing what I've proposed at the top of my 2009, which is why people looked at me like.

Jon Radoff: Well, you were way ahead of your time. I think maybe back then people did not envision just how rapidly natural language, just as one dimension of this comes along. So in the last three years, I did a graph up on my blog a few months ago and we'll throw that up on the screen for everybody here. But we've gone up like a million percent in the last three years in terms of the number of parameters that go into these deep learning algorithms. What does that mean? And what we've gone from natural learning algorithms that are really bad and you can just immediately tell it's a computer to now it can actually write stuff where you're not too sure whether it came from a human or not. In fact, you know, and a lot of times sometimes it's uncanny valley. Sometimes it's pretty convincing. Like you can play a game like AI dungeon, for example. You can tell that there's some weirdness about it part of the time. But it's pretty interesting. And that's just GPT-3, which is the current version. It's going to GPT-4, which is like another order of magnitude and it includes spatial logic, spatial understanding. How is that going to transform, or at least that whole trajectory going to transform this frontier of virtual beings?

Evo Heyning: I think we're going to start to see more conversations between the models. So GPT-3 and 4 represent a whole way of thinking and modeling. And there are others. So having different personalities with very different voices and very different ways of putting information together and then putting them in conversation is going to create some very interesting ways of understanding the world differently than humans would have ever come up with.

Jon Radoff: Yeah. And I get excited about combining this idea with the decentralization we were talking about earlier, because there can be more than one world, too.

Evo Heyning: It ought to be possible to have a virtual being

Jon Radoff: that plays with me, I don't know, is probably against the terms of service, but plays with me in Fortnite, and is my buddy in there. But also goes to a completely other world with me. And the virtual being has its own identity, existence, intelligence, relationship with me, and my friends, independent of the particular game or world or whatever we're doing.

Evo Heyning: That's beautiful, the familiar, and that concept of the familiar that can go with you. I love that idea, and I can see where that would be extremely valuable, both as a way to store memory, right? Because that's the other part that's not portable right now. Our communities aren't very portable and our social graph, but also our memories. We're stuck to these screenshots or video captures and things like that. And so I think there is absolutely a role for the familiar, the pet that comes across the metaverse and maybe unlock doors for you where you can't unlock them

Evo Heyning: yourself.

Evo Heyning: I think we start to see that fictionalized and go see free guy because I think you'll enjoy some of those. That plays out. There are these utilities that are being developed to basically solve some of these problems from an infrastructure systems point of view. But we have to remember what AI is really good at, collecting lots of data, making sense of it, having recognition, there are certain things. AI needs to be entrusted to hold for us so that we can do the hard stuff and trust each other to do that hard stuff together, right?

Evo Heyning: The relationship building, right?

Evo Heyning: If we free ourselves up of some of these layers of noise, we can actually do more on the interpersonal and the creative. But we have to figure out how to sort of hand off and delegate the right things so that we can focus in on those important things.

Evo Heyning: Yeah, and you brought back portable communities.

Jon Radoff: It's an idea I want to maybe get out there a little bit more for everyone. So first of all, I think that you can see a version of this inside certain wild gardens. So if I look at Roblox, for example, I think one of the things that's super powerful about Roblox granted, it's got walls around it and the game's experiences. You have there don't go outside of Roblox. But there's a lot that they've gotten right in terms of just making the offering process super easy. One of the other things, though, is the social relationships, the kids hanging out with each other, your friend group in there, you can be playing one experience in Roblox in a moment

Evo Heyning: and just be like, OK, I'm done with this.

Jon Radoff: Let's go do something else. And that's a virtually frictionless process to go from that one experience in Roblox to another. And for people that aren't super familiar with Roblox, the biggest misunderstanding is Roblox is not a game. Roblox is a whole environment for making games and experiences and stuff. And each one kind of stand on its own and people make stuff there. But I think that's super powerful just to observe how even inside the wild garden environment, how powerful it is to be able to either have the community or maybe not the quote unquote community that has sort of this ongoing identity. But even the sort of that immediate community, the group of 10 people that you just met in the game there, you're often doing something else. When people can do that online, I think that's going to be online, meaning outside of the wild garden. I find that to be really tantalizing in terms of the possibilities for launching new product games, experiences that are not so caught up in these company towns where they're going to take a big part of it and try to control everything around you.

Evo Heyning: I think we're going to continue to see more and more approaches to that problem. In our last interoperability demo night, for example, the team at Immerspace, they'll say was sharing how they're approaching it on the WebXR and Open Web side. So basically like chess clubs. But then you can go to different chess rooms and find your friends. And then let's go to the Dojo and play instead of to the Alice in Wonderland and play. So by doing that to the Open Web with your social graph

Evo Heyning: as part of this next generation community, almost like a meetup.

Evo Heyning: I do think we're going to start to see more and more of this lightweight and simple things happening on the Web as well as in these more curated and cultivated platforms like Roblox. Now, I love bouncing around in Roblox every once in a while. And I love that my friends in Roblox are ages nine to 70. The thing that really impressed me, that surprised me most, was I meet a lot of kids there, but I also meet a lot of parents and grandparents there. And I think because of the nature of the way I create. So for those of you who haven't used Roblox, there's a studio creator tool inside of it. And it's like a solar system where you can make your own planets. So like Magrathia and the Hitchhiker sky. It's like you can just go and make your own experiences or games or whatever you want for that community. So I made a Twin Peaks theme park, right? And people who show up at that are very different.

Jon Radoff: I know.

Evo Heyning: I know. I'm from the roller coasters and some people like Twin Peaks.

Jon Radoff: And some people are very confused. What is this thing I ended up in?

Evo Heyning: Exactly. It's a good indoctrination. And honestly, Twin Peaks is a really good analogy for portals and the metaphors, because it is a fictionalized version of that. So it's kind of a nice way to begin talking about walking through a portal and going from one experience to the next.

Jon Radoff: Now, Eva, as we're coming up on our hour here, I just want to give you a chance to maybe what is your big idea that you really want either Hammer Home, Reinforce, Share with everyone here so that they can bring it into their thinking and maybe let to challenge them about the metaphors in this future of the internet that we're stepping into.

Evo Heyning: The metaphors is what we created to be. It can be the sum total of all of our dreams and all that we want to see in the future. It can be the protopian, idealized space we want to create in reality. And it can be the place where we test out those ideas. So whether you're interested in games or work, digital twins, enterprise collaboration, just creative play and artistry, music, theater, all of it, finds a place there. And right now is a great time to experiment with how you can create value with others. Get your own ideas out and maybe do some really interesting experiments, do a little R&D on your own and test yourself. But also think about how you can connect with others, collaborate and maybe make something that is bigger than all of us. I'm excited to take all of those stories, those documentaries. I believe we're going to see a whole new boom in media from the metaphors and that it needs to be owned by all of us. So that's where all of you as creators fit in because we are the media, we are the value. It's not held by a separate tech company, it's not in a server, it's in us. So we own it, we get to choose right now to take ownership of that and create it to be what we want it to be.

Jon Radoff: Yeah, I love it. That's a beautiful thought. I hope it comes true. That's what I'm working to make come true. I mean, I see a future where, number one, you just have to make it super easy to create this stuff. And I think Roblox had the same idea, just make it easy to create stuff and let people create experiences and worlds and characters and all these things. But next we have to make it easy, but also able to control your destiny. Right? To decide what to do.

Evo Heyning: You work in mapping. I want to pull more into in the future. And this is something we're looking at at OpenMetiver Central operability is how do we begin to make robust evolutionary maps of our fields so that we can see how these things are connected so that we can do some real needs assessment around interoperability, the gap spaces that we need to address. And then figure out how to collaborate and come together. So any of you, you're welcome to join us at OpenMetiver Centraloperability as omigroup.org, Discord GitHub. Honestly, Discord and GitHub are the fabric of the OpenMetivers right now, and we invite you to participate in that.

Jon Radoff: Yep, and we'll include some links in the show notes and we'll splash it on the screen here with Josh's magic powers that he had things in later. So we'll get that in there. So Eva, it's been so awesome to have you here and hear all your thoughts that I hope this inspires everybody to not only dream about the future of Metivers that we're going to have, but to get involved either as a creator or to shape some of this. Like, it really is getting a lot easier to create this stuff and play a role in it, and you doesn't have to be in those company towns who we're talking about. That said, if you want to create there, that's fine too. Just get out and create. Like, I think that's the most important thing. Like, express that part of yourself. So by the way, if you love this kind of content, if you like hearing from people like Eva, then subscribe down below, and we're going to keep bringing those to you about once a week. We're going to have all kinds of thought leaders. You heard from Amy Jo Kim and Raff Costa so far. You heard from Eva today. We're going to have a lot more of these. And it's super fun conversations that I think are really going to challenge your understanding what the internet is and what it can become in this Metivers. So thanks so much for being here, Eva.

Evo Heyning: This has been a pleasure. Thank you, John.