Explore innovations in AI and blockchain with Michael Heinrich of 0G.AI. Join us for an insightful conversation with Heinrich, the chief executive officer of 0G.AI, as they delve into the transformative world of blockchain and AI at theCUBE + NYSE Wired event, Crypto Trailblazers. This video provides a comprehensive exploration of the latest advancements in layer-one AI technology and how they set to revolutionize the tech industry.
In this engaging discussion, Heinrich shares their expertise in establishing 0G.AI, a leading player in decentralized AI technology since its inception in 2023. They elaborate on the mission to create an AI ecosystem that ensures transparency and reduces the risks associated with AI black boxes. The video, hosted by industry experts from theCUBE Research, delves into topics such as AI for security, data supply chains, and the societal impact of these technological advancements.
Key takeaways from this conversation include the importance of verifiability and traceability in AI, as articulated by Heinrich, to safeguard against biases and data poisoning. The discussion also underscores significant insights into innovative solutions for scalability challenges in blockchain, according to industry analysts. Viewers gain a deeper understanding of how these advancements lead to significant societal and economic changes.
#0GAI #LayerOneAI #BlockchainTechnology #CryptoTrailblazers #ArtificialIntelligence #theCUBE #NYSEWired #TechInnovation #AIForGood
Find more SiliconANGLE news and analysis https://siliconangle.com/
Follow theCUBE's wall-to-wall event coverage https://siliconangle.com/events/
Learn about the latest theCUBE events https://www.thecube.net/
00:00 - Intro
00:06 - The Crypto Frontier: Exploring theCUBE, Layered Technologies, and 0G's Vision
02:40 - Navigating AI: Challenges in Transparency, Bias, and Accountability
05:42 - Societal Impact of AI and Decentralization
07:58 - Exploring Data Supply Chain and Blockchain Solutions
10:39 - Layer-One Infrastructure and Data Throughput Challenges
14:18 - The Vision for Scalable Decentralized AI
17:20 - Use Cases and Adoption of 0G Technology
21:00 - Decentralized Governance: Navigating Challenges and Future Innovations in AI and Blockchain
Forgot Password
Almost there!
We just sent you a verification email. Please verify your account to gain access to
theCUBE + NYSE Wired: Crypto Trailblazers. If you don’t think you received an email check your
spam folder.
Sign in to theCUBE + NYSE Wired: Crypto Trailblazers.
In order to sign in, enter the email address you used to registered for the event. Once completed, you will receive an email with a verification link. Open this link to automatically sign into the site.
Register For theCUBE + NYSE Wired: Crypto Trailblazers
Please fill out the information below. You will recieve an email with a verification link confirming your registration. Click the link to automatically sign into the site.
You’re almost there!
We just sent you a verification email. Please click the verification button in the email. Once your email address is verified, you will have full access to all event content for theCUBE + NYSE Wired: Crypto Trailblazers.
I want my badge and interests to be visible to all attendees.
Checking this box will display your presense on the attendees list, view your profile and allow other attendees to contact you via 1-1 chat. Read the Privacy Policy. At any time, you can choose to disable this preference.
Select your Interests!
add
Upload your photo
Uploading..
OR
Connect via Twitter
Connect via Linkedin
EDIT PASSWORD
Share
Forgot Password
Almost there!
We just sent you a verification email. Please verify your account to gain access to
theCUBE + NYSE Wired: Crypto Trailblazers. If you don’t think you received an email check your
spam folder.
Sign in to theCUBE + NYSE Wired: Crypto Trailblazers.
In order to sign in, enter the email address you used to registered for the event. Once completed, you will receive an email with a verification link. Open this link to automatically sign into the site.
Sign in to gain access to theCUBE + NYSE Wired: Crypto Trailblazers
Please sign in with LinkedIn to continue to theCUBE + NYSE Wired: Crypto Trailblazers. Signing in with LinkedIn ensures a professional environment.
Are you sure you want to remove access rights for this user?
Details
Manage Access
email address
Community Invitation
Austin Federa, Double Zero
Explore innovations in AI and blockchain with Michael Heinrich of 0G.AI. Join us for an insightful conversation with Heinrich, the chief executive officer of 0G.AI, as they delve into the transformative world of blockchain and AI at theCUBE + NYSE Wired event, Crypto Trailblazers. This video provides a comprehensive exploration of the latest advancements in layer-one AI technology and how they set to revolutionize the tech industry.
In this engaging discussion, Heinrich shares their expertise in establishing 0G.AI, a leading player in decentralized AI technology since its inception in 2023. They elaborate on the mission to create an AI ecosystem that ensures transparency and reduces the risks associated with AI black boxes. The video, hosted by industry experts from theCUBE Research, delves into topics such as AI for security, data supply chains, and the societal impact of these technological advancements.
Key takeaways from this conversation include the importance of verifiability and traceability in AI, as articulated by Heinrich, to safeguard against biases and data poisoning. The discussion also underscores significant insights into innovative solutions for scalability challenges in blockchain, according to industry analysts. Viewers gain a deeper understanding of how these advancements lead to significant societal and economic changes.
#0GAI #LayerOneAI #BlockchainTechnology #CryptoTrailblazers #ArtificialIntelligence #theCUBE #NYSEWired #TechInnovation #AIForGood
Find more SiliconANGLE news and analysis https://siliconangle.com/
Follow theCUBE's wall-to-wall event coverage https://siliconangle.com/events/
Learn about the latest theCUBE events https://www.thecube.net/
00:00 - Intro
00:06 - The Crypto Frontier: Exploring theCUBE, Layered Technologies, and 0G's Vision
02:40 - Navigating AI: Challenges in Transparency, Bias, and Accountability
05:42 - Societal Impact of AI and Decentralization
07:58 - Exploring Data Supply Chain and Blockchain Solutions
10:39 - Layer-One Infrastructure and Data Throughput Challenges
14:18 - The Vision for Scalable Decentralized AI
17:20 - Use Cases and Adoption of 0G Technology
21:00 - Decentralized Governance: Navigating Challenges and Future Innovations in AI and Blockchain
play_circle_outlineIntroduction of Austin Federa, co-founder of DoubleZero, during Crypto Trailblazers event.
replyShare Clip
play_circle_outlineBuilding a Collaborative Global Fiber Network: DoubleZero's Innovative Approach to Buffering and Routing Beyond Public Internet Limitations
replyShare Clip
play_circle_outlineExploring Collaboration and Tokenization: Democratizing Network Contributions in Contrast to High-Frequency Trading Solutions
>> Welcome back. I'm John Furrier, host of theCUBE, here at the NYSE CUBE Studios. Of course, we've got our Palo Alto Studios. We're here for the Crypto Trailblazers wall to wall. It's our 12th interview of the day. Gemma Allen and I are interviewing all the leaders over the Crypto Trailblazers. We're making it happen, building out the infrastructure, building out the applications, and also re-plumbing the financial system as we know it. We've got a great guest here, Austin Federa, co-founder of DoubleZero. We met the other day when Solana was in here ringing the bell as they listed their ETF on the exchange. Huge seminal moment for the industry. Austin, great to see you again.
Austin Federa
>> Yeah, good to see you again.
John Furrier
>> Yeah, I mean, we were riffing on the floor. After the bell. You get stuck at the NYSE and you can't really leave because trades are happening. So we were chatting fascinated by two things you're working on. One is you've been working with Solana and the founders for a while, but you're doing networking stuff, okay? I'm like-
Austin Federa
>> Yeah....
John Furrier
>> "Oh, networking. This is like AI, like data center involved." Explain what you're working on, what your company's doing and what your project is.
Austin Federa
>> Yeah, As you said, we're doing networking stuff. So what does that mean? At its core, DoubleZero is an effort to build a high performance global fiber network suitable for distributed systems like blockchain and centralized systems as well. For years, we've had traditional financial players running on private dedicated fiber lines, microwave, shortwave, all these different technologies that bypass the public internet. And big tech does the exact same thing. But all of crypto today runs on the public internet.
John Furrier
>> And the thing about all the congestion, tension, routing efficiencies, I totally get that. We've been living that nightmare for years in the IT world. But what I find interesting is that when you think about blockchains, it is distributed computing meets decentralization.
Austin Federa
>> Yes.
John Furrier
>> So in the AI world that NVIDIA lives in, over the past, I'd say five years, it's been hardcore, but mostly public domain knowledge interconnects.
Austin Federa
>> Oh, yeah.
John Furrier
>> If you look at all the hottest areas, co-packaged optics, HBM memory, you know what I'm talking about? So that was a key to the AI systems. So I think you're onto something with this interconnect because networking systems together will also affect the laws of physics. Packets. How are you thinking about this? Because it's not obvious to the average person that classic networking would be part of this.
Austin Federa
>> Yeah, it's a really good observation. The real story of AI workload training is yes, the GPUs are getting faster, and NVIDIA has better and better architectures and bigger and bigger chips every year, but what they've invested in a more quiet way is the interconnects between these GPUs. And this goes back to the early days of supercomputing cluster research when it's like everyone wants to build a machine that can do a petaflop or an exaflop of data of compute. But the real challenge is how do you actually load the data fast enough? Because You have a million cores and you have a system that's pushing 100 gigabits per second. That's actually very little data per core. And so AI training now is dictated as much as how quickly they can move data in and out of the GPUs is how fast the GPUs themselves actually are. And you see this in OpenAI's work in building new data centers is, yes, they're building data centers and there's a whole power grid story about where they're putting them, but they're running a huge amount of private fiber to link all their data centers, to make them as if it's one massive data center.
John Furrier
>> And take us through how that relates to decentralization.
Austin Federa
>> Yeah. So today, all of the blockchain systems that everyone uses operate on the public internet. And the public internet is a great resource for what it is built for. It's not built for global scale financial infrastructure. This is why the world's financial system does not run on the public internet, it runs on a whole parallel structure. And so our whole innovation here was to say we can build a software system that allows anyone who operates a fiber network to contribute individual links or whole graph of network links together and contribute that to the DoubleZero network. And so we have 11 contributors today that are all contributing to our global fiber network. And this is really the only fiber network that we're aware of in the world that's run by multiple independent parties. So each contributor admins their own boxes, their own systems, but it has the technology, like multicast, like deterministic routing and path dependency, that you don't see on the public internet.
John Furrier
>> It's interesting, when I look at NVIDIA, I love coming back to the AI cycle because there's a lot of advancements there, they are saying that they're getting into quantum as well. But some say not pure quantum. So the role of the GPU and AI is also going to impact crypto. So if they're the AI factory, will there be crypto factories out there from a data center perspective? And two, will the network, which becomes the OS in these connected systems, be tokenized?
Austin Federa
>> Oh yeah. You can sort of think about this as blockchain for all of its history has run on, apart from Bitcoin miners, pretty much the most generic hardware you can think of. It uses the public internet. It doesn't really care what the underlying CPU or RAM is that it's running on. It's all levels of software abstraction. And one of the core innovations that have happened over the last few years is people have begun focusing on the same type of optimizations you see in the traditional financial space. So what does this mean? FPGAs, something that have been standard in many trading firms and exchanges themselves use FPGAs for a decade is finally coming to crypto. And so we have FPGAs in line on the DoubleZero network accelerating a lot of common cryptography and data processing tasks for blockchain networks.
John Furrier
>> So you a inbound flow of better gear, better networks coming?
Austin Federa
>> Yeah. We see there's probably two or three things that have all happened at the same time. The first one is crypto is becoming part of traditional finance. These two worlds are getting much closer together. We have a much more friendly regulatory environment. And so as you have big technology firms and big trading firms that want to start operating on crypto, they look at the environment and they say, "This networking stack is a mess." When they execute a trade on NASDAQ or NYSE, they know how long it takes to get to the matching engine exactly, and they know exactly how long it takes the matching engine to make a match. In crypto, if I'm a market maker and I'm sending a trade, well, first off, where am I sending it in the world? If it's Solana, every 1.6 seconds, a new validator somewhere else in the world is building a block. So it's like trying to rush your data to the New York Stock Exchange, but the New York Stock Exchange keeps moving from New Jersey to Singapore to Tokyo, all over.
John Furrier
>> So high-frequency trading, the Flash Boys problem-
Austin Federa
>> Yes....
John Furrier
>> has to go away because you can manage the trades on, say, what, timestamp and atomic clock? Because everyone's contending in the trade because latency does matter.
Austin Federa
>> Yes, it absolutely matters.
John Furrier
>> It sounds like it's being a rethinking of what does that even mean? I mean, UDP protocol would be great for that,
Austin Federa
>> Right? So imagine the problem that Flash Boys was trying to address, which is how do we get data as quickly as possible from Chicago to New York, and then you N scale that problem. So there's over 300 data centers that just Solana validators are run in. At any given point, if you want to get a trade as quickly as possible included in the blockchain, you have to rush to one of those data centers. And that problem is too big for one company, too big for a DRW Cumberland, or Jump, or Citadel to wire up every single one of those data centers for trading. So what do they do? They cooperate. They build consortiums, they join consortiums. And that is what DoubleZero is at the end of the day, we're a consortium.
John Furrier
>> You're basically the version of what they solve for high-frequency trading to make consistency in the routes, so there's no arbitrage.
Austin Federa
>> Yes.
John Furrier
>> You're basically doing that same thing but not making the path latency standard. Is that right?
Austin Federa
>> Yeah, we're basically taking the exact same problem set and we're applying it to a world that instead of there being maybe 10 major financial markets in the world, there's now thousands of ones concurrently.
John Furrier
>> What's your secret sauce to do that?
Austin Federa
>> Well, the true secret sauce is a token. Tokens are the most powerful unit of alignment for unaligned actors that we've created since the corporate structure. And what that means in this case are our structure allows for independent fiber contributors to all come in and contribute fiber to the network, and they're rewarded based on the proportional value they bring. So we use something called Shapley values, which is a basic economic formula to say, how much worse would this system be if we take this piece of it away? And so we can ascribe values to each link brought into the system there and they're rewarded in tokens. And that token reward comes from fees that validators and traders pay to use the DoubleZero network.
John Furrier
>> So let me zoom out for a second because I'm get my tokens crossed because there's many at NVIDIA. They have tokens , that's for AI.
Austin Federa
>> Different tokens.
John Furrier
>> In your business model, are you tokenizing network contribution or are you tokenizing the traffic and-
Austin Federa
>> Yeah.
John Furrier
>> Or both? Well, let's just say I own a fiber network.
Austin Federa
>> Yes.
John Furrier
>> I can join and donate my fiber, and I can tokenize my network and you can measure that and reward me?
Austin Federa
>> Yeah. So let's take a real example of this, Galaxy. So Galaxy is a large digital asset firm. They have a fiber network that they use, they market make, they have an OTC desk. They do all these sorts of other businesses around this as well. And so what they've gone out and done is they've taken a slice of their network and a few new links they've taken and they're contributing those, effectively donating them, to the DoubleZero network. But in exchange for that donation, they now earn tokens. And so those token rewards are directly proportional to how useful that network link is. So if there's a bunch of validators using that network link-
John Furrier
>> It's like QoS.
Austin Federa
>> Yeah. And it beats the public internet, they get rewarded.
John Furrier
>> So essentially they're just waiting to be used, not worrying about... So just managing their quality.
Austin Federa
>> Yeah.
John Furrier
>> So at any given time from a policy standpoint, it's like, I want shortest path, I want whatever, that comes in and then gets rewarded. Is there differentiated services around that? Can you say, "I'm going to reward you more for X, Y or Z"?
Austin Federa
>> There will be, yes. We haven't built this out yet, but we will have the same type of... So the whole point of the DoubleZero network is it's a multi-party network, both from the provider standpoint and the user standpoint. So the validators, you can think of them as trading venues. The validators need to be accessible through the network, but the thing that actually pays a lot of the fees are traders that are looking to get an edge over one another. And so you'll be able to do things like pay for data prioritization as your data moves over it. But we're building these as open networks that anyone can join and contribute to.
John Furrier
>> Open source always wins. I wish we had more time because we're running late. We'll definitely get you back in. You're in New York City, so we'll have you in. But first tell the origination story. How did you get here? So you've been working with the Solana team for how many years? Five years?
Austin Federa
>> Yeah. I joined Solana at the end of 2020, worked for Labs first and then the foundation where I ran strategy. And through the part of building a network like Solana, one of the main things we kept coming back to is the limiting factor. There's certainly code improvements and software improvements, but the quality of the internet came to be a major limiting factor in the scalability of this technology.
John Furrier
>> And this idea came about when?
Austin Federa
>> This idea-
John Furrier
>> In term of this idea happened. You guys sitting there-
Austin Federa
>> Like two years ago.
John Furrier
>> Yeah. Because, yeah, we're talking when the founders came in, it's like they're technologists.
Austin Federa
>> Yes.
John Furrier
>> They're like, "We want to solve hard problems."
Austin Federa
>> Oh yeah.
John Furrier
>> Were you sitting around the table drinking beer, drinking wine, coding? What was happening? What happened? Take us through the-
Austin Federa
>> Yeah. So there was this real technology problem of how do you synchronize 6,000 nodes globally as quickly as possible without putting a massive burden on each individual validator? And so there were ideas we were throwing around that were like, do we just buy every validator a 10 gigabit internet connection? But then you still have the problem of jitter and latency. That only solves your throughput issue. And so really what this came down to... And it's important to say DoubleZero is not a Solana only protocol. We'll support all sorts of high performance distributed systems. But Solana is the most pressing use case in crypto today because there's validator clients now that can do 100,000 transactions per second on Solana, but the public internet doesn't really support that at the moment.
John Furrier
>> Yeah. To your point about NVIDIA feeding data into the GPU.
Austin Federa
>> Exactly, exactly. And so what we really set out to do was build a network that had the type of security and censorship resistance that crypto would need in order to trust it. Most of us think of the internet as this big decentralized system. It's actually run by about 10 to 15 companies at the end of the day. So we need at least that many providers who can all work with each other to deal with a fiber cut .
John Furrier
>> It's a meritocracy too. If you've got a good network, you get more tokens out.
Austin Federa
>> Exactly.
John Furrier
>> All right, so what do you guys optimize it for? Put a plug-in for the effort, you guys looking to hire? Give us some stats. How many people are working on the project?
Austin Federa
>> Yeah, we've got-
John Furrier
>> Bankroll, how much cash is involved?
Austin Federa
>> We've got about 30 people working on the project at the moment, and we've got 11 independent network contributors. We're always looking for more and welcoming more folks. And then we're running about 35% of the Solana network today, which equates to about $28 billion of value.
John Furrier
>> And how do you make money?
Austin Federa
>> So validators pay a seat fee to operate on the network, and then traders and RPC servers also pay fees.
John Furrier
>> And You're collecting that, so you have revenue?
Austin Federa
>> So that money actually goes out to... I run the DoubleZero Foundation, just like the Solana Foundation or the Ethereum Foundation, we're non-profits. And so the money that comes in actually goes to the fiber contributors themselves because they have large CapEx and OpEx costs.
John Furrier
>> So they get the payouts, too?
Austin Federa
>> Right. They're the ones building fiber cables. We're just building software.
John Furrier
>> So it's a light organization, so you don't need a lot of funding.?
Austin Federa
>> No, I mean-
John Furrier
>> You're perpetual.
Austin Federa
>> Yeah, yeah. Exactly. So we have raised, we have a good treasury, but the main purpose of that is to support the advancement, adoption, security, and decentralization of the network.
John Furrier
>> Awesome. And what's next for you? What's the big to-do item?
Austin Federa
>> Yeah, so the biggest thing we're rolling out soon, well soon meaning next year, is this prioritized data, which is really suited for traders that are very latency-sensitive. And then expanding out beyond Solana to support new blockchain.
John Furrier
>> It reminds me of the old days of internetworking. It's like the same kind of things, QoS, route management, SLAs, policy, programmability, but smarter.
Austin Federa
>> But it's a whole new world. The beautiful thing about this is we are primarily not doing computer science. We're doing engineering. We're doing just work. There's 30 years of incredibly robust private networks that we can just build off of that technology stack. And it's sort of like the cloud was not a technology transformation, it was a business model transformation.
John Furrier
>> You're bringing an open network approach to essentially provisioning routes.
Austin Federa
>> Yes.
John Furrier
>> And then putting a little bit of a spin on it that's tailored to the needs on the transaction volume.
Austin Federa
>> Yeah, and this is a real example. And the SEC agreed, right? We have a no action letter for the SEC for our token model components. But this is a real model where it does not work without tokens. This would not work as a traditional equity company with billing.
John Furrier
>> Awesome. Thanks for coming in. Great to see you again. Glad we can get it in there.
Austin Federa
>> Thank you.
John Furrier
>> DoubleZero Network Foundation. Again, this is, again, the ultimate democratization of networking. And then with the quality and the communities involved, check it out. I'm John Furrier, bringing you all the action here at the NYSE Cube Studios, part of the Crypto Trailblazer series that we're putting on. Of course, it's celebrating the celebration of crypto. Thanks for watching.
>> Welcome back. I'm John Furrier, host of theCUBE, here at the NYSE CUBE Studios. Of course, we've got our Palo Alto Studios. We're here for the Crypto Trailblazers wall to wall. It's our 12th interview of the day. Gemma Allen and I are interviewing all the leaders over the Crypto Trailblazers. We're making it happen, building out the infrastructure, building out the applications, and also re-plumbing the financial system as we know it. We've got a great guest here, Austin Federa, co-founder of DoubleZero. We met the other day when Solana was in here ringing the bell as they listed their ETF on the exchange. Huge seminal moment for the industry. Austin, great to see you again.
Austin Federa
>> Yeah, good to see you again.
John Furrier
>> Yeah, I mean, we were riffing on the floor. After the bell. You get stuck at the NYSE and you can't really leave because trades are happening. So we were chatting fascinated by two things you're working on. One is you've been working with Solana and the founders for a while, but you're doing networking stuff, okay? I'm like-
Austin Federa
>> Yeah....
John Furrier
>> "Oh, networking. This is like AI, like data center involved." Explain what you're working on, what your company's doing and what your project is.
Austin Federa
>> Yeah, As you said, we're doing networking stuff. So what does that mean? At its core, DoubleZero is an effort to build a high performance global fiber network suitable for distributed systems like blockchain and centralized systems as well. For years, we've had traditional financial players running on private dedicated fiber lines, microwave, shortwave, all these different technologies that bypass the public internet. And big tech does the exact same thing. But all of crypto today runs on the public internet.
John Furrier
>> And the thing about all the congestion, tension, routing efficiencies, I totally get that. We've been living that nightmare for years in the IT world. But what I find interesting is that when you think about blockchains, it is distributed computing meets decentralization.
Austin Federa
>> Yes.
John Furrier
>> So in the AI world that NVIDIA lives in, over the past, I'd say five years, it's been hardcore, but mostly public domain knowledge interconnects.
Austin Federa
>> Oh, yeah.
John Furrier
>> If you look at all the hottest areas, co-packaged optics, HBM memory, you know what I'm talking about? So that was a key to the AI systems. So I think you're onto something with this interconnect because networking systems together will also affect the laws of physics. Packets. How are you thinking about this? Because it's not obvious to the average person that classic networking would be part of this.
Austin Federa
>> Yeah, it's a really good observation. The real story of AI workload training is yes, the GPUs are getting faster, and NVIDIA has better and better architectures and bigger and bigger chips every year, but what they've invested in a more quiet way is the interconnects between these GPUs. And this goes back to the early days of supercomputing cluster research when it's like everyone wants to build a machine that can do a petaflop or an exaflop of data of compute. But the real challenge is how do you actually load the data fast enough? Because You have a million cores and you have a system that's pushing 100 gigabits per second. That's actually very little data per core. And so AI training now is dictated as much as how quickly they can move data in and out of the GPUs is how fast the GPUs themselves actually are. And you see this in OpenAI's work in building new data centers is, yes, they're building data centers and there's a whole power grid story about where they're putting them, but they're running a huge amount of private fiber to link all their data centers, to make them as if it's one massive data center.
John Furrier
>> And take us through how that relates to decentralization.
Austin Federa
>> Yeah. So today, all of the blockchain systems that everyone uses operate on the public internet. And the public internet is a great resource for what it is built for. It's not built for global scale financial infrastructure. This is why the world's financial system does not run on the public internet, it runs on a whole parallel structure. And so our whole innovation here was to say we can build a software system that allows anyone who operates a fiber network to contribute individual links or whole graph of network links together and contribute that to the DoubleZero network. And so we have 11 contributors today that are all contributing to our global fiber network. And this is really the only fiber network that we're aware of in the world that's run by multiple independent parties. So each contributor admins their own boxes, their own systems, but it has the technology, like multicast, like deterministic routing and path dependency, that you don't see on the public internet.
John Furrier
>> It's interesting, when I look at NVIDIA, I love coming back to the AI cycle because there's a lot of advancements there, they are saying that they're getting into quantum as well. But some say not pure quantum. So the role of the GPU and AI is also going to impact crypto. So if they're the AI factory, will there be crypto factories out there from a data center perspective? And two, will the network, which becomes the OS in these connected systems, be tokenized?
Austin Federa
>> Oh yeah. You can sort of think about this as blockchain for all of its history has run on, apart from Bitcoin miners, pretty much the most generic hardware you can think of. It uses the public internet. It doesn't really care what the underlying CPU or RAM is that it's running on. It's all levels of software abstraction. And one of the core innovations that have happened over the last few years is people have begun focusing on the same type of optimizations you see in the traditional financial space. So what does this mean? FPGAs, something that have been standard in many trading firms and exchanges themselves use FPGAs for a decade is finally coming to crypto. And so we have FPGAs in line on the DoubleZero network accelerating a lot of common cryptography and data processing tasks for blockchain networks.
John Furrier
>> So you a inbound flow of better gear, better networks coming?
Austin Federa
>> Yeah. We see there's probably two or three things that have all happened at the same time. The first one is crypto is becoming part of traditional finance. These two worlds are getting much closer together. We have a much more friendly regulatory environment. And so as you have big technology firms and big trading firms that want to start operating on crypto, they look at the environment and they say, "This networking stack is a mess." When they execute a trade on NASDAQ or NYSE, they know how long it takes to get to the matching engine exactly, and they know exactly how long it takes the matching engine to make a match. In crypto, if I'm a market maker and I'm sending a trade, well, first off, where am I sending it in the world? If it's Solana, every 1.6 seconds, a new validator somewhere else in the world is building a block. So it's like trying to rush your data to the New York Stock Exchange, but the New York Stock Exchange keeps moving from New Jersey to Singapore to Tokyo, all over.
John Furrier
>> So high-frequency trading, the Flash Boys problem-
Austin Federa
>> Yes....
John Furrier
>> has to go away because you can manage the trades on, say, what, timestamp and atomic clock? Because everyone's contending in the trade because latency does matter.
Austin Federa
>> Yes, it absolutely matters.
John Furrier
>> It sounds like it's being a rethinking of what does that even mean? I mean, UDP protocol would be great for that,
Austin Federa
>> Right? So imagine the problem that Flash Boys was trying to address, which is how do we get data as quickly as possible from Chicago to New York, and then you N scale that problem. So there's over 300 data centers that just Solana validators are run in. At any given point, if you want to get a trade as quickly as possible included in the blockchain, you have to rush to one of those data centers. And that problem is too big for one company, too big for a DRW Cumberland, or Jump, or Citadel to wire up every single one of those data centers for trading. So what do they do? They cooperate. They build consortiums, they join consortiums. And that is what DoubleZero is at the end of the day, we're a consortium.
John Furrier
>> You're basically the version of what they solve for high-frequency trading to make consistency in the routes, so there's no arbitrage.
Austin Federa
>> Yes.
John Furrier
>> You're basically doing that same thing but not making the path latency standard. Is that right?
Austin Federa
>> Yeah, we're basically taking the exact same problem set and we're applying it to a world that instead of there being maybe 10 major financial markets in the world, there's now thousands of ones concurrently.
John Furrier
>> What's your secret sauce to do that?
Austin Federa
>> Well, the true secret sauce is a token. Tokens are the most powerful unit of alignment for unaligned actors that we've created since the corporate structure. And what that means in this case are our structure allows for independent fiber contributors to all come in and contribute fiber to the network, and they're rewarded based on the proportional value they bring. So we use something called Shapley values, which is a basic economic formula to say, how much worse would this system be if we take this piece of it away? And so we can ascribe values to each link brought into the system there and they're rewarded in tokens. And that token reward comes from fees that validators and traders pay to use the DoubleZero network.
John Furrier
>> So let me zoom out for a second because I'm get my tokens crossed because there's many at NVIDIA. They have tokens , that's for AI.
Austin Federa
>> Different tokens.
John Furrier
>> In your business model, are you tokenizing network contribution or are you tokenizing the traffic and-
Austin Federa
>> Yeah.
John Furrier
>> Or both? Well, let's just say I own a fiber network.
Austin Federa
>> Yes.
John Furrier
>> I can join and donate my fiber, and I can tokenize my network and you can measure that and reward me?
Austin Federa
>> Yeah. So let's take a real example of this, Galaxy. So Galaxy is a large digital asset firm. They have a fiber network that they use, they market make, they have an OTC desk. They do all these sorts of other businesses around this as well. And so what they've gone out and done is they've taken a slice of their network and a few new links they've taken and they're contributing those, effectively donating them, to the DoubleZero network. But in exchange for that donation, they now earn tokens. And so those token rewards are directly proportional to how useful that network link is. So if there's a bunch of validators using that network link-
John Furrier
>> It's like QoS.
Austin Federa
>> Yeah. And it beats the public internet, they get rewarded.
John Furrier
>> So essentially they're just waiting to be used, not worrying about... So just managing their quality.
Austin Federa
>> Yeah.
John Furrier
>> So at any given time from a policy standpoint, it's like, I want shortest path, I want whatever, that comes in and then gets rewarded. Is there differentiated services around that? Can you say, "I'm going to reward you more for X, Y or Z"?
Austin Federa
>> There will be, yes. We haven't built this out yet, but we will have the same type of... So the whole point of the DoubleZero network is it's a multi-party network, both from the provider standpoint and the user standpoint. So the validators, you can think of them as trading venues. The validators need to be accessible through the network, but the thing that actually pays a lot of the fees are traders that are looking to get an edge over one another. And so you'll be able to do things like pay for data prioritization as your data moves over it. But we're building these as open networks that anyone can join and contribute to.
John Furrier
>> Open source always wins. I wish we had more time because we're running late. We'll definitely get you back in. You're in New York City, so we'll have you in. But first tell the origination story. How did you get here? So you've been working with the Solana team for how many years? Five years?
Austin Federa
>> Yeah. I joined Solana at the end of 2020, worked for Labs first and then the foundation where I ran strategy. And through the part of building a network like Solana, one of the main things we kept coming back to is the limiting factor. There's certainly code improvements and software improvements, but the quality of the internet came to be a major limiting factor in the scalability of this technology.
John Furrier
>> And this idea came about when?
Austin Federa
>> This idea-
John Furrier
>> In term of this idea happened. You guys sitting there-
Austin Federa
>> Like two years ago.
John Furrier
>> Yeah. Because, yeah, we're talking when the founders came in, it's like they're technologists.
Austin Federa
>> Yes.
John Furrier
>> They're like, "We want to solve hard problems."
Austin Federa
>> Oh yeah.
John Furrier
>> Were you sitting around the table drinking beer, drinking wine, coding? What was happening? What happened? Take us through the-
Austin Federa
>> Yeah. So there was this real technology problem of how do you synchronize 6,000 nodes globally as quickly as possible without putting a massive burden on each individual validator? And so there were ideas we were throwing around that were like, do we just buy every validator a 10 gigabit internet connection? But then you still have the problem of jitter and latency. That only solves your throughput issue. And so really what this came down to... And it's important to say DoubleZero is not a Solana only protocol. We'll support all sorts of high performance distributed systems. But Solana is the most pressing use case in crypto today because there's validator clients now that can do 100,000 transactions per second on Solana, but the public internet doesn't really support that at the moment.
John Furrier
>> Yeah. To your point about NVIDIA feeding data into the GPU.
Austin Federa
>> Exactly, exactly. And so what we really set out to do was build a network that had the type of security and censorship resistance that crypto would need in order to trust it. Most of us think of the internet as this big decentralized system. It's actually run by about 10 to 15 companies at the end of the day. So we need at least that many providers who can all work with each other to deal with a fiber cut .
John Furrier
>> It's a meritocracy too. If you've got a good network, you get more tokens out.
Austin Federa
>> Exactly.
John Furrier
>> All right, so what do you guys optimize it for? Put a plug-in for the effort, you guys looking to hire? Give us some stats. How many people are working on the project?
Austin Federa
>> Yeah, we've got-
John Furrier
>> Bankroll, how much cash is involved?
Austin Federa
>> We've got about 30 people working on the project at the moment, and we've got 11 independent network contributors. We're always looking for more and welcoming more folks. And then we're running about 35% of the Solana network today, which equates to about $28 billion of value.
John Furrier
>> And how do you make money?
Austin Federa
>> So validators pay a seat fee to operate on the network, and then traders and RPC servers also pay fees.
John Furrier
>> And You're collecting that, so you have revenue?
Austin Federa
>> So that money actually goes out to... I run the DoubleZero Foundation, just like the Solana Foundation or the Ethereum Foundation, we're non-profits. And so the money that comes in actually goes to the fiber contributors themselves because they have large CapEx and OpEx costs.
John Furrier
>> So they get the payouts, too?
Austin Federa
>> Right. They're the ones building fiber cables. We're just building software.
John Furrier
>> So it's a light organization, so you don't need a lot of funding.?
Austin Federa
>> No, I mean-
John Furrier
>> You're perpetual.
Austin Federa
>> Yeah, yeah. Exactly. So we have raised, we have a good treasury, but the main purpose of that is to support the advancement, adoption, security, and decentralization of the network.
John Furrier
>> Awesome. And what's next for you? What's the big to-do item?
Austin Federa
>> Yeah, so the biggest thing we're rolling out soon, well soon meaning next year, is this prioritized data, which is really suited for traders that are very latency-sensitive. And then expanding out beyond Solana to support new blockchain.
John Furrier
>> It reminds me of the old days of internetworking. It's like the same kind of things, QoS, route management, SLAs, policy, programmability, but smarter.
Austin Federa
>> But it's a whole new world. The beautiful thing about this is we are primarily not doing computer science. We're doing engineering. We're doing just work. There's 30 years of incredibly robust private networks that we can just build off of that technology stack. And it's sort of like the cloud was not a technology transformation, it was a business model transformation.
John Furrier
>> You're bringing an open network approach to essentially provisioning routes.
Austin Federa
>> Yes.
John Furrier
>> And then putting a little bit of a spin on it that's tailored to the needs on the transaction volume.
Austin Federa
>> Yeah, and this is a real example. And the SEC agreed, right? We have a no action letter for the SEC for our token model components. But this is a real model where it does not work without tokens. This would not work as a traditional equity company with billing.
John Furrier
>> Awesome. Thanks for coming in. Great to see you again. Glad we can get it in there.
Austin Federa
>> Thank you.
John Furrier
>> DoubleZero Network Foundation. Again, this is, again, the ultimate democratization of networking. And then with the quality and the communities involved, check it out. I'm John Furrier, bringing you all the action here at the NYSE Cube Studios, part of the Crypto Trailblazer series that we're putting on. Of course, it's celebrating the celebration of crypto. Thanks for watching.
John Furrier