Chainfeeds Introduction:
Biconomy member Nishil analyzed the representative projects of the three major application types of AI agents in cryptocurrency, as well as the two major challenges currently faced.
Source:
https://www.biconomy.io/post/ai-agents-transforming-the-web3-user-experience
Article author:
Nishil
Viewpoint:
Nishil: The evolving field of AI agents can be roughly divided into the following categories based on the type of use case and the added value that the protocol brings: 1) Autonomous portfolios, where the AI agent aims to maximize the returns of the asset pool by delegating assets to various DeFi strategies by leveraging the right set of off-chain AI data streams; 2) Prompt-based AI agents, where prompts may become the preferred way for ordinary users to interact with the blockchain in the future; 3) Gaming agents, where AI agents are designed to act on behalf of users or as pets/assistants in games. These agents are designed so that they can also interact with other agents and trade assets. AI agents face two major challenges: 1) They cannot be truly autonomous: Currently, AI agents can recommend on-chain actions and prepare transactions for users, but they still require user signatures or approvals; 2) If they choose automation, they lose security: Protocols tend to explore alternative solutions to automation, such as approving transactions, centralized vaults, shared private key pairs, etc., which make the protocol the custodian of user assets and introduce significant risks. We need to provide protection for AI in the form of user-defined permissions, that is, strictly define which actions the AI can sign and which actions it does not have the authority to sign. 【Original text in English】
Content Sources 
