Story introduces experimental framework for AI agents to interact onchain

Story protocol, an IP blockchain network, has introduced a theoretical “agent-to-agent” framework allowing artificial intelligence agents to trade intellectual property with one another onchain. 

In a Dec. 17 post to X, Story unveiled its Agent Transaction Control Protocol for Intellectual Property (ATCP/IP), saying it would allow AI agents to IP such as “exchange training data [...] investment strategies and more.”

Introducing Agent TCP/IP:

An experimental framework that lets agents trade IP with each other.

Agents can now exchange their training data, creative style, investment strategies, and more, ushering in an intelligence economy. 👇 pic.twitter.com/9NZFpKQy9T

— Story (꧁IP꧂) (@StoryProtocol) December 16, 2024

In its white paper, Story said its ATCP/IP system addressed some of the limitations currently placed on AI agents, specifically allowing the agents to trade things contained under the broad umbrella of intellectual property. 

“Currently, AI agents can only perform simple transactions, including buying and selling tokens or interacting with different wallets, but not complex agreements and contracts.”

“The inherent nature of AI agents revolves around intellectual property (IP) – from the data sets agents are trained on to the unique outputs they produce. These assets are at risk of being scraped by AI systems without credit or compensation,” said Story in a statement shared with Cointelegraph. 

Story said its ACTP/IP creates a system to manage this IP securely, allowing autonomous exchange between agents where the bots are compensated for the exchange of data. 

“All contracts executed on this protocol are ironclad, legally binding, and enforceable, both on and off-chain.”

Story said their framework would integrate with a range of major AI agents, including ai16z DAO’s Eliza, Zerebo’s ZerePy, Crossmint’s GOAT, and Virtuals Protocol’s GAME. 

The protocol added that an experimental demonstration of the framework would be released on Dec. 18, showing AI agents trading with one another using the ACTP/IP system. 

Related: Google touts ‘new agentic era’ with AI agents powered by Gemini 2.0

The framework marks one of the first steps toward a future imagined by Ethereum co-founder Vitalik Buterin. 

In November, Buterin outlined the concept of information finance, where AI technology and autonomous agents could be used to trade prediction markets with much higher degrees of efficiency than their human counterparts, potentially resulting in more accurate predictions about the future. 

AI agents, AI-themed memecoins soaring

AI agents and AI-generated memecoins have rapidly become one of the biggest emerging narratives in the crypto industry, with AI-themed tokens collectively bootstrapping more than $3 billion in market capitalization since October, with the sector now worth a total of $6.9 billion.

The top AI meme tokens by “mindshare” on X include ai16z (AI16Z), Fartcoin (FARTCOIN), Goatseus Maximus (GOAT), Zerebro (ZEREBRO), and aixbt by Virtuals (AIXBT), according to data from Kaito.

Several analysts have pinned the fusion of artificial intelligence and blockchain technology to transform Web3 and usher in a world where autonomous AI agents freely interact with humans on Web3.

Others have warned AI could introduce catastrophic risks to crypto, including more sophisticated scams, and provide increasingly advanced technology to criminals. 

Magazine: Influencers shilling memecoin scams face severe legal consequences