Bringing IP Onchain With Story Protocol | Andrea & Jason
4 Feb 2025 · 1:01:42
Jason Zhao and Andrea of Story Protocol join 0xResearch to discuss how the project brings intellectual property onchain and validates ownership claims. They cover Story's primary use cases, its Agent TCP/IP framework for AI agents to interact with IP, and the challenges of exporting IP across chains. The conversation closes with a look at centralized competitors and where onchain IP infrastructure is headed.
Speakers
Topics
Transcript
right now people don't do the right thing because it's it's the more complicated way to do it right if I want to legally license something it's it's a pain in the ass it's it's it's a big friction point if I can just click a button and have it licensed I'm much more inclined to do it because I know I'm going to save myself a ton of headache nothing set on zerox Research is a recommendation to buy or sell Securities or tokens this podcast is for informational purposes only and any views expressed by anyone on the show are solely our opinions not Financial advice pachio Ryan and our guests May hold positions in the company's funds or projects discussed what's up guys welcome back to another episode of ZX research today I'm joined by Jason and Andrea from uh story protocol thanks for coming on guys thanks for having us
excited for this um before we get started do you guys want to give a background on what you guys are building over at story yeah definitely so um I'm Jason one of the co-founders of story Andrea is our head of devell and uh you might know him as de Alias on Twitter so this is yeah he he does look like a Roman Emperor but not quite as stoic um in terms of what we're building at story so story is a purpose built layer one and we're very very focused on doing something I think blockchains haven't done for a while which is bring an entirely new type of as class on chain so our our belief as story is that you know the killer application of blockchains is creating Unstoppable programmable digital markets and um obviously Bitcoin did that uh by turning you know energy into sort of store of
value uh ethereum salana base a lot of the sort of core architecture of blockchains now is designed to be basically create uh payments Network or or currency Network um and what we've done is kind of reimagine what a blockchain looks like from the ground up if it were made for basically a very valuable you know $61 trillion intellectual property asset class right so digital property rights for ideas whether that's scientific research whether that's media whether that's the world's largest Brands we want to bring that on chain and make every single piece of Ip every single piece of media that you might see on Twitter or on Netflix or um that a farma company might interact with programmable in this Universal ledger so at a high level story is a peer-to-peer intellectual property system and um it has two highle components if you think about it
conceptually um we can go into the tech breakdown later but conceptually there's two components one is a universal Ledger for IP so you can think about you know how complex intellectual property is right now where you know companies actually spent a lot of money trying to figure out what IP they own before they can even transact with it right so it's a very a liquid but highly valuable asset class that has really really high transaction cost and a lot of those transaction costs come because there you know are centralized intermediaries that don't have standardized uh formats for registering IP so with the US copyright office or patent office might uh store IP as will look very different from the way that a Pharma company does it and will look very different from the way that Disney does it and so to do any transactions across those domains
becomes very expensive and across countries just like with payments it's it's very very expensive and so to create a standardized Global programmable Ledger for IP is step one and then the second concept that story does is because every asset has specific monetization terms so you can imagine you know I can use for example this character for 10 usdc with 50% rev share all of that is embedded uh into the interface every single asset has its own sort of smart contract interface and it's embedded into the smart contract State itself so now you can actually create these really interesting markets for IP That Couldn't exist before where you don't have to rely on a centralized intermediary or teams of lawyers or you know tons of Licensing contracts to create a deal you can actually just you know with a single API call um trade exchange license use monetize IP across
any application that's that's plugged into story so our goal is to be the basic IP layer of not just like blockchains but actually much bigger than the entire internet and we have worked with you know Partners like stable diffusion that that aren't in this order of crypto World um and and you know AI is obviously a huge use case for us because I think IP as a concept is is traditionally is getting killed by Ai and it has no chance of survival so I think the only chance for IP to really exist is in this sort of uh programmable blockchainbased Market format and I think like the largest AI providers like stable diffusion starting to adopt that where you know every output of staple diffusion can be registered on story but even more importantly like if you're an open AI you can actually use story as
you know the world's largest training set because you can make sure that you can get you know all this IP coming from millions of applications in a single click so that's what we're trying to do is is build a layer one that's purpose built for IP um and build this peer-to-peer IP system fantastic Andre do you have anything you want to add or I mean honestly I think uh I I'll just try to distill what Jason just said just as to like reiterate it but every IP is now an API right and AI can now use IP confidently knowing exactly what terms are there and they can do that programmatically um so you can if you were to consider the Corpus of potential
data as you know um imagine them being like programmable States you can condition your search on of what data to train on based on what is sufficiently licensable um versus you know asking for forgiveness later which is the default model that models use which is they gobble up everything and then if someone is powerful enough to challenge them they'll pay out a a fine uh instead of actually licensing that data from the start um and so we really want to change this model where it's IP holder first AI model is definitely a necessary part of evolution uh Innovation we're not trying to stop that
we're actually trying to encourage it but encouraging it in a way that's sustainable that makes sense um so I want to kind of take it a bit from I guess the beginning when I am dealing with a centralized party let's say I want to how would that process work um I have I hold IP that I want to sell versus like from a very basic overview level how would it work on story protocol yeah happy to run through that and I can you know I'm going to anonymize the examples here because I don't want to expose any of our partners but um let's just say I'm a centralized IP company right and what I mean by just using general example because it could be you know um something like a a magazine right that sells its IP or its
brand it could be like a Hollywood studio it could be a Pharma company but let's just assume like this generic company that most of the revenue comes from IP so what I've heard in many different instances from many different industries of Ip is that let's say I hold IP someone comes to me right so first of all they have to find a way to reach out to me there's no like standardized Marketplace there's it's not like you know trading on you know a DEX or coin base um you have to go and have some sort of warm intro you get connected into the organization somehow that introduction makes its way to me as an exec of this company and the first thing I need to do is I need to actually go spend a lot of money in my own legal team's time to figure out whether I own
the rights in question and as absurd as that sounds I've heard this many times where deals are not done with IP because the cost of actually discovering what slice of the rights that company owns is more than the deal being proposed so the revenue is not worth the cost which is absurd because if your whole business is based on IP you'd assume that it is very easy and very simple to know what IP you want but that is actually not the the base case today um so then I would spend some time maybe months figuring out what I owned maybe thousands of dollars if not tens of thousands of dollars um often more to do that and then I would have to negotiate with this person who may may or may not still be interested right so now we have to go uh you know
engage in this series of negotiations we bring our lawyers into both sides there's business considerations legal considerations there's a set of terms that is basically opaque to everyone except for the two parties involved and then there's a deal struck um if if we're lucky uh and that deal needs to be constantly audited and enforced over a period of you know months or years however long that deal lasts um and if something goes wrong then of course there needs to be enforcement action that will take you know at the very least one year and tens of thousand dollars on both sides right so that's kind of how IP works today which is which means that number one it is um extremely inefficient extremely heterogenous extremely illegible um and I think it it means that functionally only the top players with a lot of capital or a lot of Ip can participate
in that market so the entire longtail of that market doesn't exist today um so it's almost like if you could you know only if the world's largest companies could trade um stocks right or or if only like the biggest hedge funds could trade right you're you're kind of like Robin Hood users or or Dex users for example would not be able to participate in that market and so that's the state of the world today on story what that would look like is if I'm an IP catalog or an IP holder if I have IPS that I want to be um licensed or I want to generate revenue from I can set my terms for what I think that IP is worth so I might say I think this is worth you know if I'm a very large IP catalog I might say this is worth a million dollars
right I might say this is worth 1 million usdc you have to pay that upfront and also share 10% of any upside that you generate with my IP so those terms are now browsable on story either on chain or through front ends that are built on top of story and someone who's interested can go browse the universe of IP on story find the right asset if they find my asset then they can decide hey this is too expensive or this is perfect they can you know click a button obviously we need to make sure that the payment has been made on chain but once all those conditions have been satisfied then I would actually have a license which is also a token to use this IP on chain and if revenue is generated um you know there's two ways that Revenue can flow back to the original holder one if
it's onchain obviously we can see that and there's essentially a wallet address that's generated for each deal um that then routes all the revenue back to the original Ip holder and if that IP by the way if that original Ip holder has other commitments right if for example they owe 50% of their revenue to some parent company that can also flow programmatically right so it's an endtoend system that we built on stories so it's not necessarily just one to one it can be one to many or this big graph of payments um but then all of that would be on chain all that would be permissionless and it would be facilitated through you know any applications that build like great front ends on top of story so that process is basically a single click a single API call as Andrea put it um without the
need for any intermediaries other than the software itself which I think should be the goal of every blockchain is is you know the software is intermediary um you don't need a bank you just have this Dex smart contract or amm smart contract and that's your intermediary so that's kind of what um the vision of story is is to build this peer-to-peer way for people to transact uh intellectual property as an asset fantastic and how would I um how does the how does story validate an IP asset how does it determine that this is like a valid IP asset to trade or for me to be able to put it on chain and create it yeah Andrea do you want to talk about kind of the the sort of work stream that you've been you've been driving with story orchestration service and the IP validation service and how we
approach that sure so um just to take taking a step back you know story is a decentralized ledger for for IP and um can't rely on a single entity to determine whether something is is legitimate or not otherwise that would stifle it it would create a funnel that would basically replicate the same problem that we're trying to solve um so what we've come up with is a form of a decentralized signaling system uh which we're calling the ipvs the IP validation service and in its vzero form um it's mainly made up of attestations so we're working with different service providers for different types of media fil and these service providers will
send uh whenever an asset is registered especially when an asset has commercial terms attached meaning there is potential Revenue to be made um it will automatically ingest the uh the file these service providers will analyze it and will send their attestation of whether it's infringing it's potentially infringing uh where the IP was registered from so if I'm using for example an an application where everything is handdrawn right uh the attestation that this IP comes from that application can carry some weight because you can prove that it was handmade not that it's not infringing right because I could be handdrawing a picture of a famous character or something but um there's different layers of what a valid IP is right it's
like okay who was it made by um what does it represent um and so different sets of service providers will will attest to different layers of of this legitimacy um and the idea is that over time this network becomes uh you know vast so it's has a lot of redundancy it has a lot of different service providers and um and in true decentralized fashion you know is is uh the aggregate of all of these signals make up whether an IP is legitimate or not and then just to build on that so this is like the attestation layer so the decentralized network of signals um the story protocol itself has a dispute module so um if if you see an IP on on on story that
uh you know you want to dispute you you can do that and there's an arbitration process it's made in a modular way so you can hook into different arbitration processes um and we're also working with some uh pretty creative ecosystem projects uh like vario that are um building reaking on IP So based on where you allocate your basically your where you stake your funds represents how much confidence there is in and that will also tie back into the signal providers that I mentioned earlier um so I I think the tldr of all of this is we're trying to we're building it in a way that it's decentralized uh but that the aggregate result is a reliable indicator of
something um being genuine and licensable um and and then that's not even getting into the the portal which we can probably get into later yeah and just to add one thing um I think Andrea covered it super well it is important to distinguish why this is necessary right because IP in data are sort of similar but not exactly the same and I think IP is much more valuable and broader as a concept and the reason why is because especially you know I think AI training is a good lens to view this distinction if I training a model I can have 10 terabytes of crap data right like that's very possible and I'm going to have a very garbage output right the saying in AI is like garbage in garbage out so I can spend as much computer as I want I can have the best model the best you
know uh time reasoning whatever Chain of Thought um and if I have terrible rep you know repeated data or just lowquality data um maybe it's like the same website scrapes like over and over again I might have a lot of data but it's going to be very uh it's gonna it's not going to be very valuable right and so I think the idea with stories like everything is unique everything is validated everything is pristine and that's why it's actually so valuable for AI companies to partner with us they're not worried about just getting 10 terabytes of of of noise they're actually going to get 10 terabytes of signal right and so that work that we do is is important I think that separates us from like the average you know deepin style thing where it's just collect information sort of in this halfhazard way we have a very strict standard for
what constitutes IP because it is it has to be sort of like original unique idea and we we do make efforts to to scale our qualification of the inputs into story system that makes sense and um I guess like to talk a bit about use cases um what do you guys see as the primary use cases for story prot how would um what does the typical customer look like for story yeah so I can speak at a high level and I think Andrea can dive into some of the interesting AI use cases that we've seen I think that's definitely the major one but in general the way that our product works is um it's not necessarily a single application so uh the the best way I like to frame story is similar to stripe
where stripe isn't saying hey if you want to buy a hoodie on store a and you want to buy like uh I don't know like a hockey stick on store B um you go to stripe.com and check out right like the idea is that stripe is building this developer platform where you can easily integrate stripe checkout into uh hundred different types of stores right and the and the goal of stripe is not to be taking this stealing the show from all these stores it's actually just be in the background providing like this essential service right so without stripe you can't pay obviously stores need to have a checkout but like it's not you don't go to any store for stripe story operates in a similar way where you know we're building this blockchain on the back ends but also on the sort of front end side we've built a really you
know robust set of apis and sdks that allow we already have you know 50 to 100 ecosystem apps some of them are web two applications that have you know millions of users um others are crypto native and they can actually use our API right um as they would use a stripe API it doesn't have to be smart contract level um Integrations and essentially plug in stories IP service whether they want to supply IP into stories so a lot of apps use us like hey we have a bunch of creators we want to register their IP on story so they can monetize on 50 other apps right because it's a peer-to-peer system others say hey I have you know a comic creation tool or I have like some sort of scientific research platform and I want to pull all this interesting stuff from story and expose it to my
user so both supply and demand are possible some apps do both if you're a social media app you probably have uh both supply and demand within the same app um but that's the way we operate and in terms of use cases I think you know we've seen a lot of traction in AI which I think Andrea can cover also outside of that I mean IP is like a very valuable asset class people kind of because people don't get to engage with IP they don't really think about it but it's a 6 I mean it's actually more but I think in 2023 was a61 trillion asset class according to the world intellectual property organization and so we've already started to see you know really interesting like ip5 defi use cases come up where for example last week Justin Bieber's peaches like that entire song um was on ramped to story as an rwa so
all the yield will be on chain all the rights are on chain and actually in the ecosystem project that did that was called Arya they they um raised a $7 million vault in N9 minutes so was almost $1 $1 million per minute it um that people are just so excited about being able to access this asset class right and so now imagine like you can take out a bond on Mario right like Nintendo can like Source financing from their IP it's like a wild idea but I mean it's obviously on chain where people are going to experiment with this first or fractionalizing IP IP packs me coins I think that world is pretty interesting we're starting to see that ecosystem develop but also to me like the thing that I'm most excited to BU is is AI um so Andre could talk about that one last thought is as we get into 2025
clearly the uh the tide has changed institutions are coming and they are and governments are here so one thing that story I think is really well suited to do is to engage with government offices so we've been I've been talking to my friends over in Washington DC about how we can kind of integrate with maybe the US you know patent office Trademark Office copyright office um or even with other governments and I think story can be like a much more efficient backend for these governments and also standardize IP across the world so that's also kind of it's not necessarily A use case that exists right now but it's something that we're really pushing hard on um but yeah Andrea do you want to mention the AI side of things yeah for sure um so you know uh everyone who hasn't lived under a bush what's the
saying uh a rock a rock or a bush this is where the Italian part comes up where I just completely mess up the the sayings um but is you know a AI especially AI agents have become pretty much uh The Talk of the Town right now and one of the first things that we noticed as we were Building Story is agents and models really are pretty much pure IP both in terms of what data goes into them right which is all IP especially high value High signal data and all of the content and knowhow and skills that come out of these models right so if if an agent is truly autonomous and can conduct tasks um for example be an expert Trader right um other agents might want to license and
borrow those skills uh to conduct specific trades right and so both the inputs and outputs are IP and there's going to be uh right now it's arguably and this a little bit of a hot take most agents are just thin wrappers around llms like you know open Ai and thropic but we see things stirring you know bubbling from from the surface and you know there's been a big volcano eruption a couple days ago with the Deep seek uh you know just going crazy and um basically disrupting the whole AI world because finally there's an an open- Source model that is cheaper to train by 40x times and pretty much anyone that has a video
card can now fine-tune that model and create purpose built agents right and the the reason I focus on purpose-built agents is because as agents become purpose-built and it's it's not just a prompt that differentiates them from one another but it's actually a custombuilt model that no one else has done done potentially built on data that no one else has access to right think of um medical research think of um Finance algorithms right not just web 3 use cases and as these Mo as these purpose buil agents become more and more autonomous other agents will want to transact with them and so we actually wrote a white paper um a different white paper called Agent T tcpip which is transaction control uh protocol for
intellectual property uh because we really believe that this interchange of Ip of knowhow is going to be a central part of the agentic future and we actually have agents already using it so one of the examples that uh most recently uh re really uh you know really found super interesting is uh Da Vinci so so Da Vinci is is an agent built on story that is acting as an actual agent like in the sense of a how an actor has an agent right so an agent's agent and so it already negotiated rights with Luna the the main agent from virtuals protocol to represent her uh when it comes to uh you know licensing her her
IP to others so so so Da Vinci is now Luna's agent and they're using agent tcpip to transact with one another and to um basically negotiate rights and um you know we're also collaborating with a lot of different AI framework works we have a agent tcpip um integration with uh with Eliza and we're collaborating very closely with the Eliza team we also are co-sponsoring hackathon e Denver um we um we are working with the zerra team like I think every single agent team that's that's currently in the space with the Talis AI team uh there there's there's a lot of innovation happening uh I just before this I just got off a call
with with a team I I'll try to stay as vague as possible because I I don't think it's announced yet but uh they're working at at a pretty big important innovation layer which is the memory systems for agents right because um again going back to the point about agents being purpose built and being uh uh and and being unique everything hinges on them having memory and being able to accumulate as much memory as possible and so they're really innovating on that and they want to use story natively within that protocol to be able to register uh for example chain of thought data for example U you know specific fine-tuning data as IP that other models can license and incorporate into their own models um so it really
seems like a lot of these Ai and agent teams are really thinking about the IP layer and and I think that's why we've been having so many conversations and it's good to see because you know the default in the last two years since since uh chat GPT came out was let's just scrape all the data and then see what happens and now people are waking up and saying okay we actually need to make sure that there's you know some uh checks in place and and IPS and IP Control and and but adding control and adding attribution without adding friction that's the core unlock right because you can definitely go the old way but it'll just grind everything to a halt but but being able to keep the IP
layer the licensing layer without any of that friction I think is is is the real appeal for a lot of these teams yeah and just just one last thing is we actually you know we had this partnership with stability AI so I think it just goes beyond just agents in the crypto worlds like to me one thing that's unique about story is I think decentralized AI is interesting right but I think where we are now is let's take like an existing stack uh in AI layer of the stack in AI like computes or like storage or or inference and let's try to decentralize it which which is interesting to crypto people but actually raises the cost and complexity for most users who who may not be ideologically interested in this right so what story is doing and why we're able to partner with a company
like stable diffusion is I think that this like peer-to-peer IP system is something that just doesn't exist today and it's something that is very needed and every single major AI company is trying to improve their IP relations their creator relations and every single AI company does not want to think about lawsuits but they are embroiled in many many lawsuits and so for us to be able to provide you know what maybe one of the world's largest training sets and they can pick and choose programmatically what they want uh by quering on chain I think that's actually a big unlock for them so um I think you know our our goal is there's a lot of interesting Asian experimentation on chain but also more importantly like being able to reach out to these large players and and embed with them and offer this new stack I think that's also
a priority for us and very exciting perfect I want to get into the um agent tcpip and then also more into AI use cases but before that I have two questions that I've been thinking about one is um what happens in the case of private IP can there be such a thing get like I guess more private intellectual property that I don't want like published on the blockchain but I do want to sell um how would that work out yeah so that's a really important question because I think it I think a lot of our more institutional Partners don't want everything that they have to be immediately publicly available so this is something we're actively working on the way that our uh most of the IP is stored on story right now is through
decentralized storage like R we and ipfs so it is public um and the thesis behind that is to start uh with the IPS that are most open and amenable to being composed on top of so the way that I think about it is you know if Mr Beast published a video on YouTube he probably wouldn't complain that everyone can see it right the whole point is you want people to build on top of it but that's not the case for everyone um and so one thing that we're actively researching on our end with the research team is you know to what extent number one can we actually investigate new types of native storage on story that does allow for Access Control in a much seem more seamless way than let's say ipfs R weave uh and then two is how do we actually leverage ZK in all this is there a way
for example not everyone's going to want to migrate to onchain storage regardless of whether it's story or some other solution so like if your stuff is on S3 like can we actually use ZK to make sure that number one it is what you're saying it is but also number two that um it's only revealed to people who meet the terms right so that's something that I think is super important that we haven't fully implemented but we are like actively researching and I think it'll be a combination of building our own file storage or or having some layer on top and then and then this sort of ZK privacy preserving um Pro programmatic way of proving uh that IP does exist and that it you know that the payment terms have been metant what's up guys um welcome back to zerox research today I'm joined by Antonio from dydx thanks for
coming on Antonio hey thanks so much for having me dydx V4 is nearly at 250 Bill cumulative volume which is huge what um what was the thesis behind V4 and are you happy with it so far yeah so the biggest thesis behind V4 is is decentralization so dydx has been around for quite a while now especially in crypto terms over seven years very much OG in the defi space um and we've had a great evolution of dydx but with dydx V3 so the one older version of dydx that was something that was built on a layer two in partnership with starkware and dydx has always been non-custodial you know transparent but it hasn't always
had this aspect of full decentralization which we consider really quar Mission and is obviously a big part of what it means to be in Defi and with V4 all aspects of the system are fully decentralized so it runs on a cosmos SDK based blockchain that we developed more or less from scratch um runs on many independent validators and the whole system from the validators to the Oracle providers to everything else is fully decentralized and yeah I have been happy with it so far um it is and it has been a challenge just developing in full decentralization world um it just makes everything relatively more complicated to build right like it's it's you have to have a higher degree of when you're
rolling something out to the blockchain itself is it scalable is it secure all these kind of things which we take very seriously um but I think we are really getting our feet under us now in terms of having a quality product in V4 and also having a really good amount of development velocity there and we're really just aiming to build something that is competitive with even centralized exchanges from a ux perspective but at the same time is fully decentralized is controlled by the community and even more than controlled by the community it's controlled by you the user and that's really what we're aiming to do here and with products like the mega vaults just one click you're the liquidity provider with instant Market listings one click again list
whatever you want it's really just designs to give you the power to trade whatever you want fantastic and you mentioned that uh there there's a few other changes that you want to implement into the product to make it the best um per derivatives product out there what are some of these of changes yeah so we're really focused on just the next few months right now but in those few months the biggest thing that we're focused on is a lot of quality improvements both to the product to the underlying infrastructure making it the most performance giving consistency to the order book which is something Traders have been asking for um and making a number improvements to the front end to things like onboarding um so right now it takes on the order of
minutes to deposit to dydx where about to release something where you can deposit in seconds from any chain making the deposit flow a lot more reliable and these are just things that we've heard as huge feedback from our users and we're really responding to that the other thing that's coming pretty soon from a product perspective that I'm personally really excited about is a much simpler version of dydx um and it still is exactly the same protocol under the hoods but one of my big thesis has been hey can we expand the amount of users that are trading perpetuals by an order of magnitude or two and I think it's actually really possible to create a relatively simpler trading experience that still is every bit is powerful
still has all the same features but just is targeted at a relatively more retail type of user and it's not all the way down the retail Spectrum it's not your friend who bought their first Bitcoin yesterday but it is something where the current perpetuals products whether they're centralized or decentralized are really complicated like if you go on a dydx or hyper liquid or a binance all the products pretty much look the same it's like there's tons of order types there's lots of mumbo jumbo about leverage and tick size and all this kind of stuff um and funding rates and it just gets pretty confusing so can we take that down really streamline the user experience make it mobile first and make that accessible to just many more users and I think that's a really
worthwhile experiment to do and something I am really excited about and that'll be coming soon all right let's get into the agent TCP IP I I read over this um I get it I definitely get it but can you guys explain it one more time about how it's going to work it's like I did my homework but I'm kidding I'm kidding um yeah so I'm I'm going to try and keep it as as uh as eli5 as possible so um um when it comes to transferring something between two parties there's two types of assets right um if I send you 10 usdc you can do whatever the hell you want with those 10 usdc right if I
uh send you um even a house if I sell you a house you you can do whatever you want with the house because it's now completely your property when it comes to IP things are a little bit more complex because I could give you access access to my IP under certain conditions under certain terms right for a limited time uh and I expect something back for example a percentage of Revenue share an upfront uh cost um and so being so just having pure transactions standard transactions is not enough to represent that right so think of it as like a transaction on steroids and agent tcpip is trying to um well actually outlines a a standardized model around which this
transactional steroids can happen and the the very core principle of it is two parties in this case two agents negotiate on terms and the way that they can do it efficiently is that these terms are programmable right and easy to reason in for uh for AI agents and we actually have like a prepackaged training data that agents can take and and absorb and they'll automatically like in The Matrix be upskilled to to to know the programmable terms um once they negotiate on a specific set of programmable terms then the terms are codified and a license is minted and so there's there's trace of what was agreed
on and uh by who and um if there's any payment required the payment occurs but that's like an optional detail and then the data is transacted on um or passed over based on those C certain conditions now the reason why story especially lends itself well well to to agents and this is another potentially hot take that that Jason may not agree with but um uh is I think agents are better suited for story than humans uh because if you think about Licensing in general uh you unless you have like you know unless you're really good at remembering everything and remembering oh I licensed 10 different pieces of Ip and these were the different terms even
if I programmatically queried all of them and I know that they're pretty much similar I a human I still need to remember especially if I'm an independent Creator what the terms are you know um what the what the limitations of my usage are the cool thing about agents is that they can they excel in absorbing a ton of knowledge and keeping track of it right and so imagine a swarm of Agents transacting with one another and uh transacting on on IP at different levels of the hierarchy so like I'm licensing a remix of someone else's IP and they can keep track of millions of different terms effortlessly um and and and so it it that that's not a limiting factor right
um I think I think this is where um even on The Human Side I think more and more people will lean on agents to help them with this sort of with with these sort of tasks so even if the licensing aspect is completely streamlined with with story I still might want to use an agent because it's going to simplify remembering in what what would I need to abide by remembering when I you know when the when the yearly schedule to pay my royalties is um and so we're also working on building like a story agent to to act as a um as like a as someone that you can delegate your IP related tasks to um and not just humans but also agents themselves so
agents will have the option of training themselves on it and and transacting directly or leaning on the on some form of story agent and again this is like completely work in progress so um but but but that's the idea is is to have um a purpose built agent for that that makes sense and with um the agent TCP IP is this what you guys are um collaborating with Eliza and Zer BR that are on wh or is this um just part of it and there's other stuff coming in as well I'm basically trying to ask is like so agents can um use the tcpip but additionally they can just also use story protocol in
general as well right like the contributors can take data from story protocol yeah I I'll also let let let you chime in but um the the way I would frame agent tcpip in the whole realm of story is Agent tcpip outlines a specific way in which an agent to agent transaction around IP can take place but a large part part of what happens in the transaction is a story related operation which is registering IP if it doesn't already exist uh associating uh specific terms and minting an agreement on those terms right and that's what we've been building for for the last two years right um so even though agent tcpip seems like a pretty recent thing it
actually leans on a lot of uh prior work and and then story can be obviously used agent to human human to agent human to human uh and and and so I would see agent TCP as like a subset of story yeah I agree I think it's uh it's like uh story is the underlying framework and agent TCP IP is sort of an interface for agent to agent transaction story yeah so it's a little bit of a funnel um uh but it is certainly relying on story and is a subset of things that you can do on story that makes sense um for royalties and revenue how is that distributed out it also sounds like there's some parent Dynamic over there which I can you guys also expand on that because Andrea you
mentioned that if I get somebody else's Royal like if I pay for somebody's IP and then I Remix it then I can sell that and people would pay me how would that system work out yeah so with royalties just to take a step back and and understand how they're structured on story basically one of the reasons why or one of the areas where we had to do the most customization in terms of building this layer one is thinking about how economic commitments and relationships are structured between IP as opposed to let's say currency transactions um the way that IP is typically structured and this is more of an empirical Point than anything else is that they're typically structured in a graph-like way so if you try to represent them in a in a database
it kind of looks like a graph with nodes and edges and the best example I think most people understand that I can give is like think about Super Smash Brothers which is you know obviously very popular very fun game we have it out the office but basically there's you know there's Mario that you can fight with Mario Bowser but you can also fight with Sega characters like Sonic or you can bring in people from you know characters from many different franchises to to compete in Super Smash Brothers right and so if you had to represent that on story what would that look like well the game itself would be a node it would be like a circle in this graph and then each of the 30 to 50 characters would be their own node as well but then they would be linked as parents uh or or you know a
source IPS for the uh Super Smash Brothers game right so now you have this like dot with like 30 dots above it that all connect um but imagine on chain you know you can have recursive situations so each of those 30 parents can have their own parents right so maybe each of those 30 parents has um their own two parents that they're derived from and so if Super Smash Bros as a game generates Revenue uh Revenue moves in on one node of this graph but it needs to flow to every single subsequent node right and if you have this tree or this graph that is exponentially growing because at each level you have you know more and more parents um with different types of terms they're not necessarily all the same terms right so you know you might pay more for one character than for the
other that character itself may have five commitments to other IPS that you know are each differ from each other and so you have this like essentially computationally exposive graph traversal problem so that's you know a lot of the work that we've done on the engineering side is to structure that so that it is like actually very cheap and very scalable um and the way that a royalty module works is that every single IP is an atomic Sovereign unit on story and it declares how much it needs to receive in recurring revenues in order for someone to initially use it and so essentially whenever Revenue flows into that graph we simply read every step every every connection that that node has right so that node might owe uh IPA like 20% it might owe ipb 30% right and and then we just sort of distribute the funds on
chain in that way um and then you know you go one layer above and that that IP that ipb that got 20% might might have its own commitments uh Upstream of it right so that's kind of how how it works is you have the sort of almost waterfall effect where Revenue flows through the graph um no matter where it enters it will find its level essentially so that's kind of how it's structured and obviously there's details to that and then last point I'll add is you know obviously when it comes to web 2 applications we need to Fiat onramp and that's something that you know we are working on but ultimately I think that's an area where you know you you're we're actually trying to bring a lot of uh revenue on chain so I think this will be good for the space where you know
eventually if YouTube integrates like a lot of that revenue for for creators on YouTube that are using story will actually eventually flow on chain right and it may move off chain if the creators choose to but I think that's kind of a a good onboarding tactic to kind of increase the tvl of the space that makes sense um one question is let's say I um buy IP off Andrea but I I don't want to pay him um or I don't just don't have money do I need to put on some collateral to make sure that like from what you said it sounds like I need to hold some form of collateral that can be slashed or cut um in case like I don't have enough money or I miss a payment right it depends on the licensing terms depend on the terms of
the IP um so some IP and this is completely up to the owner of the IP to set um so you can set upfront if if you're afraid of use cases like that uh then you would set an upfront payment to use the IP and to to obtain a license you would have to pay right um and that's kind of like automatically enforced um for or as Jason was saying like offchain let's say you build a really cool character that's registered on story and I I fall in love with it and I want to create a t-shirt line or something right a swag line or I want to use it in some IRL use um then I would me a license for it and I would be legally obliged to share if the if the terms dictate a certain percentage back
to you right so let's say I make $100,000 in sales and I need to pay you 10% then I would need to send you 10,000 um this is where you know the enforcement if if you like there's there's no perfect way to solve enforcement um like you can't solve it in the real world uh what we but the the the terms that are created on story even though they're programmable are actually codified in a legally binding document that you know has been created with the help of several legal teams um and and so it is legally binding and legally enforceable in the real world
um much like any other form of Ip agreement so if if someone licenses it and and makes money and doesn't pay you back then you know it's the same thing as if I gave you permission to use my my logo and you made uh t-shirts and you never pay me back so um one way to look at it is story is is not trying to build walls to to to to make it like we're trying to break down walls and make it easier for honest actors that legitimately want to use IP and and and um and give back to to the creators of the IP and actually have a self-sustaining uh economy around IP do so uh similar to how you know uh we went from the default being going to stream
something or torrenting something to just watching it on on legitimate platforms because the the path of lease resistance was to do the right thing right right now people don't do the right thing because it's it's the more complicated way to do it right if I want to legally license something it's it's a pain in the ass it's it's it's a big friction point if I can just click a button and have it licensed I'm much more inclined to do it because I know I'm going to save myself a ton of headache right imagine if a ghost riter was able to uh seamlessly in one click license Drake's voice when he created um heart on my sleeve right which became like a super uh topstream song on Spotify but ended up being taken down because he didn't uh get a license
for Drake's voice even though he used an AI model to generate Drake's voice on that song um and and you know this is not Utopia like uh Grimes followed up quickly after saying if anyone wants to do something like that please do and just give me 50% so there is appetite on IP holders to to let their IP flourish but there's just no until now there's there's been no way to do that in a um in a scalable way that makes sense um one thing you mentioned Jason is onboarding capital are you guys thinking of exporting IP um to other chains for it to be traded uh I don't like I know
that you want ipy on I assume you also want ipy on story as well but is this something that you're thinking of like if there are purpose built trading chains um would it make sense to export IP to we created on those or is that something um that you're not considering so it's a good question so the way that we think about it is we our ambition is to build a core protocol for the internet in the same way that you know domain name system or HTTP no one's like a big fan of HTTP right it's just it's so prevalent that it's just used right so you don't see people wearing like I I'm a big domain name system supporter I'm so glad the domain name system exists so I think that we want to strive to that level of adoption and if that if that is be true even to 1% or
any approximation of that Vision we have to support as many different Computing environments as possible and the way that we think about it is it's not can we support other blockchains it's actually you know if you're offchain right like that that's also important most IP isn't even on chain so that's kind of the first order business um and so the way that we we build Solutions where or we're Building Solutions where if your IP is on let's say it's on optimism let's say it's on uh you know salana or or let's say it's on optos um it it's we treat it the same way as if it were offchain right essentially it's not on story so how do we bring it onto story so we basically have the same approach in both cases which is your your IP can live anywhere wherever you are we will meet you there right what we
we don't want to be a distribution platform we don't want to be like a social media platform where you have to be on story in order to get the benefits your IP is you know we meet you where you are right so for most people that's offchain but then if you're on salana you're kind of off story as well right so we kind of treat you the same way and essentially what we want to do is allow people to create intellectual property anywhere but then where it's settled and where it's being transacted is story so in the same way that you know any application uses probably any modern application is going to use five to 10 different Services um you know you can imagine the applications of the future leveraging story for their IP and then maybe they'll use another blockchain for payments maybe they'll use something
else for identity but I think story will be this sort of universal horizontal layer that scales across industry scales across and also scales across blockchains so we don't want to be sort of parochial in our view towards other blockchains but I would say we treat anything that's not on story as off story and we have a sort of standardized approach to that that makes sense um what front ends are being built on story protocol um that's something you mentioned is that there are teams working on you guys building up front ends yeah absolutely so F the first thing is we've built a sort of a public good uh it's not meant to be a killer app or anything but it's meant to sort of offer uh a central place for people to access story if it's their first time so it's called the IP portal it's going to come
out like shortly after our public main net um but the idea is you know if you don't know how to code if you are you know someone who holds IP you can easily register IP in a very consumer friendly no code way set your terms and it's now available on any application that's plugged into stories API right so 50 to 100 applications instantly for free and then also you can browse so a lot of people ask okay well what type of IP is on story you can on the portal you can browse all the IP on story and see the ter terms and the lineage the genealogy how it's related to every other IP and then also if you are an IP holder you can manage your Revenue you can collect all the revenue that you've earned all the yield that you've earned so to speak
from your assets um across the many different applications that may be using them um so it's sort of like a One-Stop shop but that's just sort of one application it's a very simple application to be honest um we see probably three or four categories of apps the first is what I would call like Creator tools so an example of this would be magma they're a fully web to application they have two and a half million creat per year 10 million IPS registered per year and it's essentially like a collaborative art making platform so it's almost like figma but for visual artists and every single asset that's registered on magma can be registered on Story by default so that's like a huge supply of assets to story and those and for magma it's great because they can they can tell their creators like we're giving you 500 not 500 yet but like you
know 50 to 100 distribution channels for your IP to be used instantly um so it's it's kind of no risk and then that's so there sort of like Creator tools consumer platforms the second is ipy so like defi applications we have applications where you can create ipb meme coins um ipw I mentioned Justin Bieber peaches we have uh fractionalization protocols things like that uh as well as lending protocols where you can use IP as collateral so um and then there's also another sort of ipy application that's pretty cool where you can actually stake on on IPS and the idea is that you talked about a bounty right it's basically other people attesting hey I believe this IP is legitimate and if it's not they get slash but if it is they can continue to earn yield right so there're sort of like economic Primitives that can
increase a signal around an IP asset in a way that the government or no centralized entity simply can right um so there's like this whole section of ipy that's really exciting and then we have a lot of AI powered applications right so this can be anything from Foundation model companies to the agents Andrea was talking about to even applications like ablo which is a sort of AI powered fashion remixing app that's been built on story they've actually partnered with Bain Balenciaga Polo um some of the worlds areg just Brands and you can use AI to kind of customize your shoe and create new shoes and so people can create new B main shoes if those are successful they get printed or made I should say and the revenue is shared with um Appo the Creator as well as a brand uh all via story right so I think there's a lot of
different um applications and we'll see a lot more continue to pop up um but our goal is you know not only incubate sort of the next generation of consumer applications um and IP related applications but also to integrate with some of the existing massive distribution channels like the Netflix YouTube twitters of the world world I mean that's ultimately our goal so that's where we want to go um with our ecosystem perfect um last question you already talked about onboarding on story so I'm not going to ask that but do you see centralized competitors um being an issue and I don't mean like centralized competitors in the real world I mean centralized competitors on the blockchain like sonum so I don't know if I'm saying that right sonum um or if
like other I guess semi IP related um blockchain pop up do you think that's how are you guys going to work on that or are you just going to assume that's like offchain as well we're just gonna take their IP yeah so I think I mean look I think what Sony is doing is really interesting I think Sony is very forward forward facing there but ultimately in I I think I kind of have the same reaction is is complimentary to story in the same way that you know I think Deutsche Bank announced that they are going to build a layer 2 on ZK sync and I don't think anyone thinks that Deutsche bank's L2 is going to replace ethereum or Orana as the settlement layer for for payments so I see it as um one instance that proves that there is interest but I don't see
it as a real competitor to our PP IP system I see it as a as a sort of traction uh that we can use and hopefully work together because ultimately you know um I don't I think our goal is to reduce the amount of essentially transaction Taxation and transaction costs that we have to suffer as a society because our ideas and our research and our creativity is basically held and arbitrated by like a few entities right so ultimately I I don't think it's going to be one of those entities that exist today that creates that system because it is utterly destructive to their business model and it it reduces their value capture by a lot right so in the same way that ethereum isn't really probably threatened by you know if Goldman Sachs started a blockchain or something like I I think uh we feel that way about you
know some of these sort of like big companies coming onto crypto but I mean obviously it's a good thing for the space so hopefully we can find a way to to work together yeah and and just to answer that I feel like a lot of these uh complimentary networks are creating IP potentially IP focused uh chains or I I IP inspired chains right like sonum obviously has access to a lot of Ip but they're less focused on creating a transacting layer for IP exchange they're more how do we bring IP on chain and so totally agree with Jason that it's a very complimentary type of approach right so like if sonian brings on a ton of Ip and start and suddenly tokenizes it a story is going to be the perfect layer on
which to um create uh terms on it to you know like to to really add that API for that specific IP that's brought on chain uh regardless of whether it's on story or not yeah I think IP IP can live anywhere and it should just be settled on story fantastic um I said that was the last question but I want to ask actually one more what where do you guys see um because IP is such a like I'll be honest I mostly do Defi and gaming stuff so IP is very far away from me so I want to get like your honest take on where do you see IP on in crypto in blockchain um over the next two years or three years like do you
think we catch on very very um quickly or do you see there being kind of a bumpy ride can I yeah can I take a a quick stab at this so actually I'm I'm GNA focus on your on your preamble to this question which is you you said you know I'm usually in defy and gaming and so IP is very far from me I think an ideal goal within the next year year or two is you become as familiar with IP um in the way that you need to be not that you need to know everything about AP but you feel it as easy and seamless to interact with as you do a def protocol or a as you do a game right because if you think about it Financial um being defi a native was a very far
concept you know 10 years ago for most people unless you worked on very specialized Industries now everyone can be a Trader everyone can be um a defi expert if they just spend 10 minutes because there's so many there's so much infrastructure making it easy M making it approachable um and we want to do the same thing for for IP not in the way not in the sense that everyone should learn everything about IP and become an IP lawyer right that's not what we're trying to do much like you don't need to know about what happens in the inner workings of of a def5 protocol but we want everyone to be a to be a user of an IP protocol right uh feel like anything that they create is good enough to be an IP right there or not it has value that's up to the
market to to kind of determine but you shouldn't think of oh my stuff is you know creating an IP is like this whole thing with lawyers and super expensive and you know I'm not going to think about it like the default Behavior should be anything you create anything that comes out of a creative mind is IP and should be registered as IP and we want to make it as easy as possible yeah I mean I think what Andrea said is interesting because IP is everywhere it just isn't it's sort of like not a construct that people have explicitly in their head so they don't really see it right because you see what you you see how you frame the world and the way you frame the world shapes that what you see and I think for example like the way I see this cycle is that the three most
exciting Trends in the space is number one Ai and agents number two it's it's meme coins and number three it's rwas right and like in every single uh vertical I see IP all over that because that's what I'm looking for right I I think that AI like IP is this core input into AI we talked about that extensively meme coins like I mean I mean that that's just extremely obvious right like with the chill guy stuff the problems are obvious too but obviously like what is a meme it's literally like a mimetic IP it's an internet native IP and then rwas you can see what we've done with Justin Bieber's peaches it's actually like this you know the problem with a lot of rwas is like why why would I want a t bill on chain I just you know it's like kind of like it's just not that
interesting right because you can access T anyone in the world can access people the entire world economy operates on American debt right so it's not like this sort of rare scarce asset whereas people are dying to get into this just a Bieber asset because like where else are you going to be able to access and democratize this thing that you know even your mom would understand why it's valuable and everyone wants to kind of financial exposure to that right so to me I see IP everywhere it's like already kind of lurking in underneath and our job is just to make sure it's explicit uh and then you know I think that the way they see story scaling is kind of like what is the other product in crypto that has true pmf it's stable coins and how do stable coins work right the way that
they work is they borrow the legitimacy of an existing institution that is a US dollar and then they make it 20x better right so staple coins are just so much easier to use than US Dollars um when it comes to International payments when it comes to um censorship resistance when it comes to being able to access like a whole programmable ecosystem of Finance Financial applications same thing with story right like it's not like we're trying to usurp the existing legal system we're borrowing the legitimacy and the power of that legal system but then we're making it 100x better by bringing it on chain and I think that's like if we want to see I think that the next few years I mean it's maybe doesn't sound too long but it is a decent amount of time and ultimately to me like uh the best way to onboard like real people
into crypto is not to improve the ux like no one actually cares about ux if you think you can yield Farm whatever 500% apy you will click all the buttons you will read all the terms and conditions you will go watch YouTube like you will you will get there right that what's what the thing that crypto lacks is it doesn't motivate people to use crypto like people are motivated to do really bad ux and jump through a lot of Hoops if they want something but like most of crypto doesn't inspire people to want something and I think what IP is is something that everyone cares about and everyone understands and so I really do see it as like the sort of intersection with traditional institutions I see it as intersection with AI so I think we'll start to see a lot more people like stability um and perhaps even
governments come in through this paino of Ip right and then we can argue about once people want to do something then we can argue about whether the ux is good enough but right now I think the problem isn't ux it's like outside of money like people don't want anything in crypto so this is bringing entirely new thing to the table that can inspire people um to jump through the Hoops that they might have to jump through to use onchain systems fantastic um guys before we head off do you want to where can my listeners um go to learn more about story and what should they do what should they be looking forward to over the next yeah so definitely the biggest Milestone is our public mainnet launch which is coming up really really soon um I can't really give a specific date but it it you know you won't have
to wait long um and we're really excited for that obviously we've been working for that for three years but for us that's just a starting line um it's not it's not the Finish it's actually just very much the beginning so there's going to be a lot of awesome applications ready on day one but also um much more exciting stuff to come and in terms of where to find us we have uh if you're a developer our docs we update them like every week thanks to Andrea and his team um so they're very very live if you have questions we have a very extensive frequently asked questions so if you're an agent we have a page for you yeah we have uh we have uh made our documentation easily consumable to agents so we have like a file that you can train your agents on uh so if you're
developer start there and then you know you can always reach out to Andrea um who who you know supports a lot of developers and then for people who are interested in in using story as an IP holder or just part of the community um our Twitter is probably the best place to get live information and we have an ecosystem page on our website uh story. Foundation ecosystem where you can start to dive into the applications fantastic uh guys thank you so much for coming on this was all yeah thanks for having us [Music]