The Integration of Web3 and AI Agents: Analysis of Emerging Narratives and Potential Projects

In the first half of this year, the total market capitalization of the crypto market grew by 37.3%. Among them, Memecoin, AI, and RWA performed the strongest. Looking at the price performance of coins from last week, AI still maintains a strong momentum, second only to Memecoin.

Web3 x AI is becoming the focus of attention for major venture capitalists, with some institutions even purchasing older projects TAO( Bittensor) in the public market. This indicates that this track is not a fleeting phenomenon but will continue to innovate alongside the development of the traditional AI field.

For example, the latest trend of AI Agents in the traditional AI field has also been introduced into the Web3 world. In the first half of the year, a large number of Web3 x AI Agent projects emerged, such as Spectral, Olas Network, etc. At the same time, some older projects like Fetch.AI and Phala are also keeping up with this narrative. At the recently concluded ETHCC, many Web3 developers and top VCs began to focus on the AI Agent narrative.

Overview of AI Agent

The AI Agent is an "agent" based on the large language model (LLM), capable of perceiving the environment, understanding autonomously, thinking independently, making decisions, and executing actions. Its core functions can be summarized in three steps: perception, planning, and action.

The main difference between AI Agent and AI chatbots like ChatGPT is:

• Chatbots are designed to interact with humans and will not take autonomous actions.
• The Agent is designed to complete autonomous tasks and possesses the ability to take autonomous actions.

For example, the AI Agent can automatically complete the entire process from information retrieval and analysis to execution of actions based on the user's needs.

The New Revolution of Web3 x AI Agent

Web3 can bring the following advantages to AI Agents:

1. Anti-censorship: Using a decentralized LLM to build AI Agents can avoid the censorship issues of centralized LLMs.

2. Decentralization/Ownership: Addressing the issue of core data being stored with centralized AI providers.

3. Monetization: Providing new profit avenues for AI Agent developers and investors.

4. Composability: Achieving interoperability between AI Agents and other Agents in the same network.

Conversely, AI Agents can also bring for Web3:

1. Conversational AI Agent on-chain, simplifying the Web3 user chain research process.

2. Implement an "intention-centered" interaction model to drive mass adoption.

Currently, Web3 x AI Agent projects can be divided into two categories:

• Web2-style on-chain dialogue AI Agent: does not include on-chain operations
• Web3 Native AI Agent: Includes on-chain operations

Spectral Project Analysis

Spectral is an AI Agent project with on-chain operation capabilities, and its main products include:

• Spectral Syntax: Officially developed collection of Agents
• Spectral Nova: Decentralized Machine Learning Inference Platform
• Agent Wallets: Cryptocurrency wallets integrated with AI Agents
• Inferchain: Layer 1 powered by AI Agent

Spectral achieves its ultimate vision through Inferchain, facilitating the development and interoperability of on-chain AI Agents.

Token Economics:

• $SPEC is used to pay for the fees of AI Agents and for decentralized governance.
• Currently circulating market value of $85 million, FDV $800 million
• Core contributors and investors will unlock starting from May 2025

Development Data:

• Registered users 65,362
• Generated contract quantity 1,055,568
• Average interaction count per user is 25
• Number of Memecoins created: 5,043

GaiaNet Project Analysis

GaiaNet is a distributed AI infrastructure that does not have on-chain operation capabilities, with main components including:

• GaiaNet Node: Individuals and enterprises can quickly deploy AI Agents.
• GaiaNet Domain: A collection of AI Agents in specific fields
• GaiaNet DAO: Governing the entire ecosystem

GaiaNet brings the Web2 AI Agent ecosystem onto the blockchain, making it easier for developers to deploy Agents and generating a collection of specialized knowledge AI Agents for various industries.

Project Progress:

• Completed a $10 million seed round financing in May 2024
• There are currently 35 available conversational AI Agents.
• Total number of nodes 18,594

Overview of Other Early Projects

• Zotto: Users can create AI Agents to execute trading intentions.
• AgentLayer: A Layer2 based on OP Stack, facilitating the interconnection of AI Agents.
• Olas Network: Off-chain AI Agent Ecosystem
• Theoriq: Modular and Composable AI Agent Foundation Layer
• AgentCoin: Universal Web3 AI Agent
• Giza: Web3 AI Agent framework, off-chain reasoning and on-chain execution

The integration of Web3 and AI Agents is rapidly developing, which may bring about entirely new user interaction paradigms and application scenarios in the future.