AI+Web3: Towers and Squares

Summary

1. AI-related Web3 projects are performing actively in the primary and secondary markets.

2. The opportunities of Web3 in the AI field are mainly reflected in: utilizing distributed incentives to coordinate long-tail supply, covering data, storage, and computation; while establishing a decentralized market for open-source models and AI agents.

3. AI is mainly applied in the Web3 industry for on-chain finance (, cryptocurrency payments, trading, data analysis ), and assisting development.

4. The value of AI + Web3 lies in their complementarity: Web3 is expected to counteract the centralization of AI, while AI is expected to facilitate the popularization of Web3.

Introduction

In the past two years, AI has developed rapidly, and the wave of generative AI triggered by ChatGPT has also impacted the Web3 field. The concept of AI has brought a wave of financing to Web3 projects, and the secondary market has become more active. Research and topics related to AI + Web3 are equally hot, ranging from AI + Depin to AI Memecoin, AI Agent, and AI DAO, with new narratives emerging one after another.

This article will explore how Web3 can play a role in various aspects of the AI technology stack, and what new opportunities AI can bring to Web3.

Opportunities of Web3 under the AI stack

Infrastructure Layer: A Shared Platform for Computing Power and Data

In terms of computing power, Web3 projects like io.net and Aethir gather idle GPU resources through a model similar to Airbnb, providing low-cost and efficient distributed computing power for AI.

In terms of data, Web3 projects like Grass and Vana acquire users' private data at a low cost through incentive mechanisms. At the same time, projects like Synesis and Sapien improve the quality of data labeling through gamification and other methods.

In terms of data privacy and security, Web3 projects such as Super Protocol and BasedAI utilize technologies like TEE, FHE, and zero-knowledge proofs to protect sensitive data.

Middleware: Model Training and Inference

Web3 proposes a decentralized open-source model market, such as Bittensor building a P2P model market, ORA introduces the initial model issuance ( IMO ) concept.

In terms of verifiable reasoning, projects like Modulus Labs utilize zkML technology to verify the correctness of AI model execution.

Application Layer: AI Agent

The decentralized nature of Web3 allows Agent systems to be more decentralized and autonomous. Projects like Virtual Protocol, Spectral, etc., provide early financing support for AI Agents through methods such as token issuance.

How AI Empowers Web3

AI and On-chain Finance

AI agents are expected to become the future of on-chain finance, applicable for information collection and prediction, asset management, and optimizing financial experiences. Projects like Bitte and Wayfinder are exploring this field.

In terms of trading security, AI can be used for real-time monitoring of unusual transactions, analyzing risks, and so on.

AI and On-Chain Infrastructure

AI can be used for on-chain data analysis, such as Web3 Analytics, MinMax AI, etc.

In the development audit area, AI can assist in smart contract development, code vulnerability checks, etc. For example, Spectral offers a one-click deployment feature, and Fuzzland uses AI to assist in auditing.

New Narratives of AI and Web3

AI brings new possibilities to Web3 fields such as NFTs, GameFi, and DAOs. Projects like Bicasso for AI-generated NFTs, AI Hero for AI-powered blockchain games, and ai16z for AI-assisted DAO governance are examples of these attempts.

Conclusion

The combination of AI and Web3 is like "towers and squares", with both complementing each other. AI brings new vitality to Web3, while Web3 is expected to alleviate the centralization risks of AI. In the future, AI + Web3 is expected to provide users with more innovative services and business models.