The AI ​​industry will move from "volume models" to "volume applications". When the model itself is no longer a scarce resource, the focus of competition will be on how to implement the model as a tool for AI agents. Therefore, the next AI Agent will take three major directions: 2 segmented application tracks + 1 big infra narrative

1) Defai, a financial service agent, including automated MEV strategy (extending MEV capture rights from miners to ordinary users), dynamic risk management (allowing Agent to automatically adjust mortgage rate, stop loss line, predict black swan), on-chain credit score (analyzing user historical behavior, personalizing DeFi interest rate and pledge rate), etc.;

2) Gamefai, a game service agent, including multi-agent collaborative mini-games (new game paradigm), human + agent gambling interaction (AI pet battle), and game NPC agentization (NPC has memory emotions and growth trajectory, and forms a virtual social relationship with players);

3) Chainization, a web2 AI Agent that connects to the "generalization and modularization" exploration of blockchain infra in the past, including decentralized computing power, data, algorithm platform, Oracle 2.0 adapted to Agent, DA storage that can dynamically manage memory, and interoperability that supports multi-agent collaborative interaction based on chain infra, etc.