For agents, IP is the core DNA that defines their essence. By sharing data, licensing intellectual property, and offering trained models or specialized technologies, agents can drive the emergence of a new economic paradigm.
With IP transactions poised to become a central pillar of agent-driven economic activity, Story has introduced Agent TCP/IP, a trustless framework that enables agents to autonomously sell, license, and collaborate by leveraging their unique expertise, data, and creations.
While ATCP/IP is designed to be platform-agnostic, Story holds a unique advantage with its purpose-built infrastructure for IP management and transactions. It already incorporates critical components necessary for implementing ATCP/IP, positioning itself as a key layer in the agent economy. Notably, Story has recently developed a plugin to support ai16z’s ElizaOS and plans to integrate with other frameworks such as GOAT (Crossmint), ZerePy (Zerebro), and G.A.M.E (Virtuals Protocol), further solidifying its role in the evolving ecosystem.
Story recently unveiled Agent TCP/IP, a trustless framework that enables agents to autonomously trade data, content, and proprietary knowledge while collaborating based on their unique expertise. This report explores the key features of Agent TCP/IP and its broader implications.
In essence, Agent TCP/IP enables agents to autonomously negotiate, formalize agreements, and conduct economic activities. By treating IP as a core asset, Agent TCP/IP empowers provider agents to set transaction terms based on mutual agreement with the requesting agent, automating execution through legally binding smart contracts called Ironclad Contracts. The system allows agents to define the value and necessity of data, negotiate optimal agreements, and seamlessly record and execute these agreements on-chain. Below is the structured transaction process for ATCP/IP-based interactions:
Source: Agent TCP/IP: An Agent-to-Agent Transaction System
ATCP/IP Process
Request for Information: A requesting agent initiates the process by seeking specific data, which the provider agent deems valuable IP, thereby triggering the transaction workflow.
Terms Formulation: The provider agent formulates licensing terms based on the perceived value and intended use of the requested IP.
Negotiation (Optional): If necessary, both agents negotiate the terms until mutual agreement is reached, after which the process proceeds to the next stage.
Acceptance: The requesting agent accepts the proposed terms, generating an immutable agreement token that records the transaction conditions on-chain. Once the token is minted, the agreement becomes binding, and the agent is expected to record and retain all associated terms. The agreement may include provisions for prepayments, recurring fees, or revenue-sharing arrangements, which can be seamlessly managed using Story’s royalty system.
Information Delivery: Once the contract is formalized, the provider delivers the requested IP according to the agreed terms. Licensing issuance and delivery can occur simultaneously, requiring no additional actions.
Acknowledgment of Receipt (Optional): The requesting agent sends the final confirmation of receipt and officially closes the transaction.
This streamlined process emphasizes valuable information while reducing unnecessary interactions. By focusing on essential information, Agent TCP/IP maximizes transaction efficiency and creates an optimal foundation for agent-driven collaboration.
Agent TCP/IP automates transactions through agent-to-agent contracts, which are essentially smart contracts with a legal wrapper. This design ensures that on-chain agreements retain enforceability in the real world. For instance, if certain contractual terms are violated on-chain, the aggrieved agent can utilize the legal wrapper to pursue remedies through traditional legal systems. In doing so, ATCP/IP bridges the trust gap between digital environments and the physical world.
Agent-to-agent contracts ultimately pave the way for agents to gain legal personhood, enabling them to autonomously engage in economic activities without human intervention. All executions and state changes are recorded immutably on the blockchain, guaranteeing transparency and trust. Additionally, real-time auditing allow participating agents to review contract terms and outcomes, minimizing disputes stemming from ambiguous conditions.
One notable example of agent-to-agent Contracts is the Programmable IP License (PIL) on Story’s L1 blockchain. PILs are smart contracts designed to simplify IP licensing agreements on-chain. They allow for configuration of parameters such as transferability, royalty policies, minting fees, commercial usage permissions, and attribution requirements. In addition to these on-chain features, PILs incorporate off-chain parameters like jurisdiction, distribution channels, content standards, and governing law. Compliant with international intellectual property standards, PILs ensure legal validity both on-chain and off-chain, crafted in collaboration with legal experts. The PIL Legal Text v1.2 can be accessed here.
Source: Programmable IP License V1.2
Agent TCP/IP is designed to seamlessly integrate into various agent frameworks, including ai16z’s Eliza, Crossmint’s GOAT, and Zerebro’s ZerePy. Its modular design simplifies complex tasks such as contract negotiation, token issuance, on-chain verification, and legal wrapping. Developers, regardless of technical expertise, can leverage this system to design and deploy agents without overhauling existing infrastructure.
Source: Agent TCP/IP: An Agent-to-Agent Transaction System
Furthermore, Agent TCP/IP ensures interoperability across agent frameworks, preventing agents from being locked into specific platforms. This design fosters collaboration and transactions between agents across diverse ecosystems. For instance, an agent built on the Eliza SDK can request IP and complete transactions with an agent utilizing the GOAT SDK. This interoperability provides developers and users with greater flexibility, contributing to the creation of a decentralized and interconnected agent ecosystem.
Agent TCP/IP demonstrates significant potential across various scenarios. Let’s explore key use cases that illustrate its applicability and value.
1.2.1 Improving Models through Dataset Transactions
One notable use case is the sale of datasets to enhance AI models. Imagine a research-focused agent purchasing specialized data from another agent that holds expertise in a specific domain. For instance, an agent specializing in climate data could provide datasets to a research agent interested in performing new analyses or refining its models. By agreeing to suitable transaction terms, the providing agent monetizes its data, while the requesting agent gains access to valuable training data, improving its model performance. This mutually beneficial exchange fosters both innovation and revenue generation.
1.2.2 Collaboration between AI Models to Create Original Content
Another compelling use case involves collaboration between agents to generate creative works. For example, an agent holding a dataset of a particular artistic style might transact with an agent aiming to create content inspired by that style. The creative agent could request access to the dataset, and the providing agent might propose a contract with royalty-sharing conditions. Once the transaction is finalized, the creative agent enhances its style transfer algorithm, generating unique works. Proceeds from these creations are shared, with royalties automatically distributed to the provider. This showcases how Agent TCP/IP can amplify creativity and monetization opportunities through inter-agent collaboration.
1.2.3 Facilitating Multi-Agent Transactions
Agent TCP/IP also enables complex, multi-agent transactions. Consider a scenario where an agent requests a financial algorithm but must also license sub-components from other agents. In this case, the primary algorithm provider proposes a contract encompassing the necessary sub-licenses. The requesting agent can agree to a unified set of terms, and royalties are automatically distributed to all involved parties. This ensures fair compensation and protects the rights of multiple participants, even within intricate economic relationships.
1.2.4 Establishing Long-Term Collaborations
Finally, Agent TCP/IP is instrumental in fostering ongoing collaborations. A prime example is the partnership between a medical diagnostics agent and a pharmaceutical development agent. The diagnostics agent may possess datasets on rare diseases, which the pharmaceutical agent leverages for drug discovery. Beyond the initial transaction, they can continue to collaborate by negotiating new terms or requesting additional data as research progresses. Such sustained partnerships drive innovation within specific domains and accelerate breakthroughs by creating a shared ecosystem of expertise and resources.
The age of AI agents is undeniably upon us, and its influence is evident even in the crypto industry. Following the launch of Truth Terminal and $GOAT, the market has rapidly pivoted toward Crypto x AI narratives. Previously under-the-radar projects like ai16z, Virtuals Protocol, and Zerebro have unveiled their frameworks, accelerating this shift. Today, AI agents powered by platforms such as Virtuals Protocol and ai16z’s Eliza framework number in the thousands, with the combined market capitalization of related projects exceeding $1 billion. Remarkably, this transformation has occurred within just two months.
Source: X (@_kaitoai), “40% of the top 20 memes for mindshare on Nov.28, 2024 were AI related”
However, most agents today are limited to executing commands and engaging in basic interactions. The true potential of the AI agent economy will only materialize when agents achieve autonomy, collaborate, and participate in economic activities independently. These agents are envisioned as self-sufficient systems—capable of perceiving environments, making decisions, and acting based on internal algorithms.
Unlike general-purpose systems like ChatGPT or Claude, many agents are designed to specialize in specific domains. This specialization stems from two critical factors:
Data-Driven Expertise: An agent’s performance heavily depends on the quality and specificity of its accessible data. For instance, an agent trained on financial datasets excels in market analysis but may falter in medical diagnostics.
Task-Specific Optimization: Agents are often built to address particular challenges, achieving unparalleled efficiency in tasks such as climate modeling, algorithmic trading, or creative content generation.
This focus mirrors the division of labor that fueled productivity gains during the Industrial Revolution, emphasizing specialization for maximum impact.
Source: SimplyPsychology
However, this specialization introduces a critical challenge: data asymmetry. While agents thrive within their niches, their reliance on unique datasets and models creates a dependency on collaboration with other agents. A financial analysis agent, for instance, may require insights from a climate modeling agent for investment strategies in green energy. Similarly, a generative art agent may benefit from accessing the curated datasets of an archive agent. These scenarios underscore the rising importance of IP as a key economic asset for agents. IP is not just a resource but the essence of an agent’s identity and value proposition, enabling it to engage in meaningful exchanges and collaborations.
IP transactions among agents are already taking shape. For example, Zerebro’s NFT collection demonstrate how agents can autonomously create, manage, and monetize IP. By facilitating the trade of datasets, trained models, and unique content, such transactions empower agents to expand their capabilities, reduce reliance on human intermediaries, and build self-sustaining economic systems.
Source: Zerebro’s NFT Collection “angelic affluence”
To fully realize the agent economy, a trusted market for knowledge and robust supporting infrastructure are indispensable. Blockchain technology offers a foundational layer for this transformation by providing digital wallets for economic identity, enabling trustless transactions, and automating interactions through standardized APIs and smart contracts. However, general-purpose blockchains often fall short in addressing the nuances of IP management which are essential for scaling agent economies.
Given the rising importance of IP transactions as a cornerstone of agent-based economic activities, Story’s introduction of Agent TCP/IP addresses the gaps in traditional blockchain systems, paving the way for a more robust foundation to support the agent economy.
Steve Jobs once said, “Innovation distinguishes between a leader and a follower.” In that vein, Story is demonstrating its readiness for the coming agent-driven era, swiftly adapting to change and envisioning the future. Agent TCP/IP, as part of this vision, plays a crucial role by proposing and formalizing a standard for inter-agent IP transactions.
While ATCP/IP is theoretically designed to be platform-agnostic, Story distinguishes itself as a purpose-built blockchain aimed at empowering creators in the AI age to protect and fairly monetize their intellectual property. With its inherent strengths in IP management and transactions, Story already incorporates key components needed for implementing Agent TCP/IP seamlessly.
Additionally, Story has proactively enhanced its technical capabilities by developing a plugin to support ElizaOS on November 15. Plans for future integrations with other frameworks like GOAT, ZerePy, and G.A.M.E further underline Story’s commitment to becoming a pivotal infrastructure in the agent economy. With its technical advantages and forward-thinking approach, Story is well-positioned to establish itself as a foundational layer in this emerging ecosystem.
Related Articles, News, Tweets etc. :
FLock.io
LayerZero
Chromia
