Abstract

Story, a peer-to-peer intellectual property network, is dedicated to creating the world's first programmable knowledge and creative market. In Story, both scientific and creative assets can be registered on a universal ledger with customizable parameters. All assets have standardized interfaces that can be used by any software application or artificial intelligence model, enabling the free flow and monetization of intellectual property. Story coordinates assets through a decentralized network graph: nodes represent independent intellectual property assets, while edges represent the legal and economic commitments of those assets. Story also features an asynchronous, decentralized verification mechanism that assesses the uniqueness of assets based on cryptoeconomic incentives, ensuring fairness and transparency. Join Story to contribute to the growth of the world's only open, permissionless repository of knowledge and creativity.

1. Introduction

The current intellectual property (IP) system heavily relies on centralized institutions to manage ownership and value circulation. From social media to pharmaceutical companies, a few unregulated entities tightly control the production and exchange of global knowledge, resulting in high costs for IP transactions. Worse still, the traditional IP system struggles to cope with the large-scale generation and distribution of content in the AI era. Works by individual creators are captured, trained, and utilized by AI without permission, leaving them without attribution or compensation; meanwhile, AI companies cannot sign licensing agreements for vast amounts of content individually, leading to a lack of transparency and liquidity in the IP market.

We need a truly open, programmable peer-to-peer intellectual property market that allows anyone to freely create, trade, and monetize without exploitation by monopolistic entities. In Story, each IP asset is embedded with programmable licensing rules that can be monetized and traded across any application. Story not only provides a universal IP repository but also offers a programmable IP market, driving exponential growth in human knowledge productivity.

2. Architecture Overview

The development of blockchain technology has gone through several stages. Bitcoin first introduced decentralized ledgers, and Ethereum built a smart contract platform on top of that, giving rise to new asset classes like NFTs. Subsequently, blockchain faced scalability challenges, leading to the emergence of high-performance public chains like Solana, which attempted to break through these limitations. Although the industry has long focused on balancing the "trilemma" of decentralization, scalability, and security in blockchain, the applicability and usability of blockchain have been overlooked, becoming a barrier to mainstream adoption of blockchain technology. Therefore, we need a new type of blockchain that not only addresses technical challenges but also creates a programmable market for complex new asset types, allowing blockchain technology to truly serve the real world.

Story is a specially designed decentralized blockchain equipped with a multi-core execution environment. It consists of a main execution core and multiple highly customizable cores. The main core is equivalent to the Ethereum Virtual Machine (EVM), allowing for the rapid adoption of existing applications within the ecosystem. The intellectual property (IP) core is one of the dedicated cores that efficiently handles the registration of intellectual property as a native asset class while optimizing the operation of complex IP relationship graphs. The IP core transforms intelligence into programmable IP assets. While Story primarily focuses on intellectual property, its flexible architecture allows for future core applications that far exceed the scope of IP-related applications.

Figure 1: The Protocol Layer of Story

Story operates on a set of decentralized nodes (also known as validators) that are connected through a peer-to-peer network. They work together to ensure consensus is reached, transactions are processed, and the integrity of the chain is maintained, building a secure decentralized network that any participant adhering to the protocol rules can join. Following the principle of separation of concerns, the protocol can be divided into three layers: execution layer, storage layer, and consensus layer.

2.1 Execution Layer

The execution layer is the computational foundation of Story, responsible for executing all transactions and smart contracts. It consists of the main core and multiple dedicated cores that work in coordination. The execution layer aims to provide maximum flexibility and computational power for each core while ensuring the stability and activity of the network.

Figure 2: Multi-Core Design of the Execution Layer

The main core is the default execution environment and is fully compatible with the EVM. The main core processes all transactions, and when a transaction calls a contract within the corresponding dedicated core, the main core can activate the relevant dedicated core. Each dedicated core's execution environment is optimized and prioritizes specific needs, such as processing speed, scalability, security, or privacy. For example, the way data is modeled, stored, and processed in certain applications can directly affect its scalability and even feasibility. The complexity of the multi-core design is invisible to users, as each dedicated core provides functionality through smart contracts, automatically activating when users interact with the relevant contracts. All cores adhere to the following requirements:

• Determinism: The execution result of any method call must be deterministic. For example, while a core may internally use a GPU to accelerate floating-point calculations, the final result must remain consistent. Non-deterministic behavior threatens the stability and activity of the network.

• Limited and Measurable Resource Consumption: Each core must accurately calculate the expected resource consumption (such as gas fees) of all contract calls and compute it within constant time based on input parameters. Incorrect measurements can undermine network stability and fair usage.

• Storage: Each stateful core can store data in key/value pairs within its storage layer's namespace, where data commitments are used for consensus. All key-value updates must follow a specified format and maintain the determinism of update order.

To meet different needs, Story's initial release deployed three dedicated cores to work alongside the main core: one core is responsible for storing and operating IP as a native asset class, another core connects the on-chain execution environment with the off-chain world, and the third cross-chain communication core helps IP assets circulate within a broader blockchain ecosystem.

Intellectual Property Core

In Story, intellectual property assets are presented through a multi-dimensional graph, where various types of nodes (such as IP assets, licenses, etc.) are connected by edges representing economic and legal relationships. These IP Graphs track the rights, licenses, and monetization processes of original and derivative assets.

Figure 3: Diagram of IP Assets

The PoC protocol deployed on the IP core is a native protocol that provides an open IP repository and a set of modules for frictionless interaction with these IPs. The PoC protocol serves as a standardized universal ledger for IP assets on Story and a means to create digital markets around these assets and their rights. The PoC protocol can trace the genealogy of IP usage, expansion, and monetization across applications, similar to how Git operates in code branching and version control. This open repository makes Story the source and exchange layer for IP assets.

IP holders can register their intellectual property on-chain and use the PoC module to customize usage terms and copyright structures. Each intellectual property can be registered on-chain as an IP asset (ERC-721) and optionally associated with an IP account (an optimized version of ERC-6551). Off-chain entities can monitor registrations, verify IP holder identities, authenticate registered IP and its licenses, and initiate decentralized dispute resolution processes in case of infringement. For more details on the PoC protocol, please refer to Appendix A.

The IP core supports the PoC protocol with an optimized data model, providing efficient local storage for IP-related assets, licenses, and certifications to optimize the processing of intellectual property relationship graphs. The IP core also offers efficient local implementations for complex graph operations (such as traversals), which would otherwise incur significant time and computational costs.

Figure 4: Flowchart of IP Registration and Usage (Standard Process Path)

For example, when a new association is established between a user and an IP asset, the protocol needs to verify compatibility by checking the IP parameters and all parent assets, ensuring the unified execution of rules. The IP core employs advanced algorithms to efficiently traverse relationship graphs and applies a universal compatibility engine for IP terms. Achieving such functionality on other Layer 1 or Layer 2 blockchains without dedicated execution cores is nearly impossible.

Off-Chain Synchronization Core

Oracles play a crucial role in the blockchain ecosystem, allowing these otherwise isolated systems to connect with real-world data and services. Without oracles, blockchains can only handle on-chain data, severely limiting their practical applications. The off-chain synchronization core provides native support for the stable and seamless integration of oracles.

Figure 5: Off-Chain Synchronization Core Facilitating Seamless Integration of Oracles and Other Data Providers

One of the services deployed on the off-chain synchronization core is Story's Orchestration Service (SOS) framework.

This service verifies the authenticity of registered IP assets and licenses, validates the social identity of creators, receives off-chain payments, and generates legally binding contracts available off-chain, extending the PoC protocol. Through the off-chain entity network, the SOS framework bridges the gap between off-chain reality and the on-chain world, providing proof for the fundamental services on Story.

Cross-Chain Communication Core

The cross-chain communication core provides the fundamental components for efficient cross-chain messaging. For example, with the Inter-Blockchain Communication (IBC) protocol, it becomes impractical to implement this protocol using EVM smart contracts due to the high computational costs of verification proofs (such as Merkle proofs) and signature verification, a problem that is addressed by the cross-chain communication core.

The ability to support cross-chain communication promotes broad interoperability within the ecosystem, eliminating value silos and expanding more application scenarios. For instance, IP holders can use their IP assets across different blockchains, accessing various markets, platforms, and services while ensuring consistency in ownership and monetization terms. Users can also use their IP assets as collateral on DeFi-focused blockchains. This interconnected approach ensures that regardless of where assets are used, the ownership of IP can be enforced without considering the execution environment, realizing the vision of Story as the IP layer for the entire network.

Scalability of Future Cores

The architecture of Story supports the addition of new dedicated cores, allowing its functionality to extend beyond the field of intellectual property. Let's look at two potential future cores.

The Artificial Intelligence (AI) core can optimize on-chain operations related to AI. By default, these operations are impractical on the EVM— as noted in paper [5], the gas cost for a simple multiplication of two 1000 x 1000 matrices would exceed 3 billion, far beyond the gas limit of a logical block. The AI core can leverage modern GPU processors for integer operations, combined with deterministic fixed-point operations, quantization techniques [6], and on-chain pseudo-random number generators to eliminate non-deterministic issues arising from randomness and floating-point errors.

The Zero-Knowledge (ZK) core can achieve efficient verification of zero-knowledge proofs (e.g., ZK-SNARK). Zero-knowledge proofs (ZK-proof) are a cryptographic method that allows the prover to demonstrate possession of certain knowledge or information to the verifier without revealing the actual information. By providing native support for zero-knowledge proof verification and utilizing a highly optimized software stack or hardware acceleration, the ZK core has advantages over smart contracts in reducing computational overhead and gas costs. This core is particularly important for applications requiring confidential IP exchanges or large-scale verifiable IP graphs. For example, it can enable IP holders to prove ownership and authenticity of IP assets without exposing sensitive information.

2.2 Storage Layer

The storage layer sits above the consensus layer, providing storage APIs to the execution layer and its cores. It simplifies the complexity of data organization, encrypting commitments to data while deciding which database to use for data management. The design of the storage layer is inspired by NAND flash translation layers (FTL), which map logical addresses to physical storage blocks to optimize write performance and extend device lifespan. The storage layer also balances performance, scalability, cost, proof size, and redundancy in heterogeneous storage systems through data placement strategies. By analyzing usage patterns and access frequencies, the storage layer can proactively optimize data distribution across different storage layers. Sub-storage systems automatically adjust based on data size, access frequency, and durability requirements to efficiently handle various data types.

Figure 6: The Storage Layer of Story

In addition to performance improvements, the storage layer addresses the fragmentation of user experience by providing a unified interface that integrates on-chain and off-chain storage solutions (such as IPFS and Arweave). Intellectual property assets come in various forms, from small text files to large multimedia files and complex databases. The cost of storing massive amounts of data on-chain is prohibitively high. When interacting with large datasets, the traditional approach is to store content on IPFS while maintaining only metadata references on-chain; this method often has significant limitations that degrade user experience. For instance, when a machine learning model is registered as an IP asset, the metadata used to capture content authenticity (such as C2PA) will be stored alongside the actual file to allow for direct model inference when needed. Notably, the underlying protocol has handled the complexity, ensuring that the behavior of the interface remains deterministic and does not compromise network stability. Technical details will be released in a subsequent technical white paper.

2.3 Consensus Layer

The consensus layer ensures the integrity, security, and stability of the network by coordinating block generation among all network participants. In recent years, significant breakthroughs and advancements have been made in blockchain consensus mechanisms. The design of this layer is highly flexible, allowing for the incorporation of new consensus mechanisms.

Figure 7: Block Formation Cycle of Tendermint (Standard Path)

Initially, Story adopts a modern implementation of Tendermint called CometBFT. Tendermint is a Byzantine Fault Tolerant (BFT) consensus engine with clear attributes and customization options, fully meeting our design requirements:

• Fault Tolerance: The system remains secure even if 33% of validators act maliciously or go offline. It ensures that all honest nodes reach consensus on submitted blocks and identifies and punishes malicious nodes through cryptographic proofs.

• Liveness Guarantee: The network can continue to progress as long as more than two-thirds of validators (measured by voting weight) remain active and participate in the network.

• Instant Finality: Transactions can achieve rapid final confirmation once packed into blocks. This instant finality is crucial for applications requiring immutability, directly impacting user experience regarding reliability.

• Modular Design: The Application Block Interface (ABCI) separates consensus from execution, ensuring that future consensus upgrades do not affect existing execution logic.

The consensus layer of Story employs a Proof of Stake (PoS) mechanism, where validators play a crucial role in maintaining the security and integrity of the blockchain. This provides economic incentives for honest and efficient operations, as malicious behavior will be punished, including the loss of staked tokens. This layer manages all operations related to staking, locking requirements, and reward distribution.

3. Application Scenarios

Beyond the technical architecture of Story, we can begin to envision various application scenarios that a peer-to-peer IP system might bring. With the emergence of a universal IP repository and a programmable smart market, we may be standing at the starting point of a new renaissance. This section will explore some of the most promising application cases, though many equally exciting applications remain unmentioned.

3.1 The Rise of Smart Markets

Bitcoin and Ethereum have redefined currency over the past few decades, but the expansion of blockchain to other new asset classes has progressed slowly. The unstoppable digital market is the "killer application" of blockchain, and a blockchain built specifically for IP will create entirely new markets.

The traditional IP system relies too heavily on complex centralized structures and intermediaries, leading to inefficiencies in the monetization of IP, a trillion-dollar asset class, in terms of readability, trading, and monetization. High transaction costs prevent knowledge and creativity from being converted into value and limit the exchange and dissemination of ideas. Story breaks these limitations by supporting peer-to-peer IP transactions and programmable IP.

Applications can integrate the IP catalogs of pharmaceutical companies or media organizations as IP RWAs (real-world assets) into Story. More excitingly, natively programmable IP can combine with DeFi applications to create an emerging field—IPFi—where IP assets can be split, used as collateral, staked, or utilized in other economically efficient ways.

Interactions between individuals are the most fundamental transactions in the new smart economy. With Story's programmable execution environment, interactions between agents and humans become possible. Agents can automatically obtain IP asset licenses from Story and upgrade their creative styles or knowledge bases through fine-tuning. If revenue is generated, agents can share part of the income with the original IP holders through Story's royalty module, creating a virtuous economic cycle that empowers IP holders with AI rather than undermining their earnings. Additionally, pure IP transactions between agents are also feasible, using Story as the settlement layer for smart agent commerce, flourishing the entire agent interaction market. We will detail this possibility in the next section.

3.2 The Foundation Layer for Artificial Intelligence

Blockchain is an ideal foundation for AI interactions, as it provides a programmable medium in which software can make hard commitments. Since agents are trained with IP as native input and generate IP as native output, Story provides a foundational settlement layer for AI transactions, suitable for both training AI models and monetizing AI outputs.

Smart Chain

In the AI field, intellectual property assets encompass various types of IP—from databases and foundational ML models to fine-tuned models and model tuning packages.

Figure 8: Diagram of AI-Related IP Assets

Figure 8 illustrates how IP assets construct a smart chain. A machine learning model (Model 2) may derive from a deep fine-tuned model (Model 1) enhanced through reinforcement learning using another database (Dataset 6). This fine-tuned deep model itself may originate from a foundational model trained on multiple databases. These training databases may also be combinations of other databases. Additionally, research on independent model fine-tuning packages is actively ongoing. Model fine-tuning software packages contain all necessary data for efficient parameter model fine-tuning, reducing the number of trainable parameters while maintaining performance close to fully fine-tuned models. Since comprehensive model fine-tuning requires substantial computational resources and memory, these packages provide an efficient alternative (i.e., Huggingface Adapters). When a model is registered as an on-chain IP asset, cutting-edge model fingerprinting technologies, such as the OML developed by the Sentient protocol [12], can capture the unique features and fundamental properties of that model.

By registering the entire smart chain on Story, economic value can be distributed among the network's nodes. Story lays the foundation for a high-quality database and model fine-tuning package market, driving the emergence of more efficient ML models. This incentive mechanism encourages data providers to contribute high-quality databases and accelerates AI development through guaranteed fair compensation. Under transparent economic incentives and automated revenue distribution mechanisms, Story becomes the negotiation, collaboration, and settlement layer for seamless cooperation among data providers, model developers, and end users, ensuring all parties have reasonable ownership.

Smart Agent Business Model

AI is evolving from independent models into an automated agent network capable of perceiving, deciding, and acting to achieve goals. This shift means that AI is no longer just a tool but is gradually becoming an ecosystem where agents create value through collaborative interactions.

Figure 9: Agent TCP/IP Protocol - Interaction Process

The Agent TCP/IP protocol released by the Story team provides a key framework for the development of automated agent ecosystems. The purpose of designing Agent TCP/IP is to promote standardized automatic interactions between AI agents. This agent-to-agent protocol lays the foundation for the exchange of IP assets (such as training data, proprietary algorithms, and creative content) between agents without relying on human intermediaries. With the Agent TCP/IP protocol, agents can negotiate, authorize, and execute agreements automatically through on-chain smart contracts, combining on-chain execution with off-chain legal enforcement, all supported by Story. The Agent TCP/IP protocol lays the groundwork for the realization of agents as independent economic entities, enabling agents to improve themselves using authorized training data and registered IP, thereby monetizing their outputs. As more agents join the network, the value and capabilities of the entire ecosystem will experience exponential growth. Each new agent brings unique capabilities and can combine with the advantages of existing agents to produce new solutions and services. For example, some specialized agents can collaboratively produce films, where a producer agent hires a writer agent and a video generation agent, distributing revenue according to their on-chain agreements.

3.3 IP Tokens

The native token of Story—IP—coordinates all value flows throughout the peer-to-peer intellectual property system.

IP serves as the underlying medium of exchange for the network, efficiently processing transactions and ensuring fair distribution of resources through a transparent fee payment mechanism whenever intellectual property assets are used, exchanged, or generate income. It not only facilitates peer-to-peer direct transactions without intermediary involvement but also acts as a functional tool and value exchange medium, aiding the growth of the network economy.

Additionally, Story employs a PoS consensus mechanism, requiring validators to stake IP to ensure network security. Validators are responsible for processing transactions and maintaining the integrity of the blockchain, receiving rewards denominated in IP based on their performance. This provides strong incentives for validators to operate honestly and efficiently, as their success is closely tied to the network's success, thereby reinforcing IP's role as the security foundation of Story.

As a unified utility token, IP serves dual functions of maintaining network security and ecosystem functionality, closely integrating network operations with the broader intellectual property market. As Story's gas, staking, and utility token, IP plays a crucial role in incentivizing all participants in the Story ecosystem, promoting the maintenance and development of the peer-to-peer intellectual property system.

In the broader ecosystem, IP is intended to play a more critical role, responsible for protecting intellectual property assets, acting as the currency for intellectual property revenue streams, and even providing a native medium of exchange for AI agents settling their intellectual property on Story. The PoC protocol and other dedicated protocols deployed on the IP core rely on IP as their operational foundation. IP facilitates all value transfers within the IP core, including royalty distributions, licensing fees, and usage-based compensation. This integration means that IP's role extends beyond basic network security—it is the cornerstone for protecting and verifying all intellectual property assets registered on Story. More importantly, the Agent TCP/IP protocol utilizes IP to provide a robust security layer, safeguarding communication between agents while offering a native mechanism for efficient value exchange between agents. This dual function ensures that AI agents can automate economic transactions and communication without the need for trust.

IP is the unified utility token for network security and ecosystem functionality, integrating network operations with the broader intellectual property market. As Story's gas, staking, and utility token, IP plays a vital role in incentivizing all participants in the Story ecosystem, maintaining and expanding the peer-to-peer intellectual property system.

Conclusion

We have launched a peer-to-peer intellectual property system that breaks the reliance on centralized institutions. The Story blockchain provides a dedicated network that can create a universal repository for intellectual property assets and a programmable market for IP asset exchange. Story's multi-core architecture offers the necessary scalability and specialization to address the evolving integration of applications and artificial intelligence scenarios. An embedded knowledge graph of intellectual property within the Story blockchain accurately depicts the complex relationships between various intellectual property assets, providing a robust record of economic commitments. As artificial intelligence accelerates, the volume and liquidity of intellectual property continue to expand, and the network built by Story will become a pillar of the new smart economy. While Bitcoin serves as digital gold, acting as a value carrier for all commodity assets, Story provides value storage for all knowledge assets in the form of programmable intellectual property.

Acknowledgments

The Story Foundation thanks a16z crypto, especially Eddy Lazzarin and Scott Kominers, for their foundational support in the early design of the Story network and the PoC protocol. Additionally, the authors would like to thank Jason Yanowitz, Chainyoda, and Scott Kominers for their feedback during the drafting of this white paper.

Appendix A - PoC Protocol

The PoC (Proof of Creativity) protocol introduces an open programmable IP protocol that elevates intellectual property (IP) as an essential component of the blockchain ecosystem. The core of this protocol is the IP asset and its associated IP account, which is a smart contract account serving as the core identity identifier for each intellectual property.

This account-centric design makes the storage and management of IP data more convenient and enables interaction and operation with this data through modular functions. These modules extend the protocol's capabilities, empowering creators and users to easily manage and license their IP in a decentralized and programmable environment.

To achieve this, the protocol consists of two main parts: data structures and modules. The data structure serves as the "noun" in the protocol, storing metadata related to IP, referred to as "IP Legos." The modules act as "verbs," providing various functionalities for IP assets registered in the protocol.

Data Structures (Nouns)

IP assets are the foundational "IP Legos" in the PoC protocol. Each IP asset represents an on-chain IP and its associated IP account, which is based on a modified ERC-6551 (token-bound account). IP assets transform a new or existing NFT into an IP entity with multiple functionalities and interactivity, representing tokenized intellectual property.

The IP account is a programmable account mapped to the tokenized IP. It modifies the ERC-6551 to include a universal "execute()" function, allowing calls to any module in the protocol through compiled byte data, enabling future module expansion.

Modules (Verbs)

Modules (i.e., verbs) are customized smart contracts used to define and extend the functionalities of IP assets within the PoC protocol.

Licensing Module

The licensing module is a core part of the protocol, allowing IP holders (as licensors) to create and manage licensing agreements using license tokens and their associated terms. The licensing terms are derived from predefined licensing templates. For derivative works, the licensing terms of the parent IP are also taken into account during the licensing process. The end result is the minting of license tokens.

The PoC protocol introduces Programmable IP Licensing (PIL) as a ready-made licensing template, aimed at providing IP holders with a universal and easy-to-adopt licensing framework, simplifying the licensing process, and establishing a standardized legal framework.

Programmable IP Licensing (PIL)

The licensing template is a legal framework compiled in a smart contract ("programmable") that specifies various licensing terms for the IP. The terms may include commercialization, transferability, or royalty percentages.

Programmable IP Licensing (PIL), developed by the Story team [14], is the first instance of this licensing template. While the terms of the IP protocol (IPA) and the issuance of license tokens occur on-chain, these contents are enforced under an off-chain legal framework. This legal contract allows tokenized IP assets to integrate into traditional legal systems and provides clear guidance for creators on how they can remix, monetize, and create derivative works.

Royalty Module

The royalty module is used to manage the revenue distribution between subordinate IP assets and their parent IP assets. It allows IP holders (as licensors) to define the revenue percentage that subordinate IP pays to the parent IP. Similarly, subordinate IP holders, as licensors, can also specify the revenue percentage that derivative IP pays to their subordinate IP.

Revenue paid to the IP asset will be deposited into the IP royalty treasury. Each IP asset has a separate royalty treasury, and royalty token holders have the right to withdraw funds from the treasury proportionally.

The royalty module supports various customizable royalty payment structures. By default, each parent IP asset can set a minimum royalty percentage. This percentage determines the share of revenue that each level of derivative IP in the derivative chain must allocate to the parent IP according to the licensing agreement.

Dispute Module

The dispute module provides users with a framework for resolving disputes. Its arbitration system consists of two main parts:

• Arbitration Policy: The arbitration policy defines the rules, processes, and responsible entities for resolving disputes.

• Arbitration Penalties: Arbitration penalties determine the consequences for IP assets that are "marked" as part of a dispute.