Written by: 0XNATALIE
Since the second half of this year, the topic of AI Agent has been gaining traction. Initially, the AI chatbot terminal of truths gained widespread attention for its humorous posts and replies on X (similar to 'Robert' on Weibo), receiving a $50,000 grant from a16z founder Marc Andreessen. Inspired by its published content, someone created the GOAT token, which surged over 10,000% in just 24 hours. The topic of AI Agent then attracted the attention of the Web3 community. Later, the first decentralized AI trading fund based on Solana, ai16z, emerged, launching the AI Agent development framework Eliza, sparking a token competition. However, the community's understanding of the AI Agent concept remains unclear: what is the core of AI Agent? How does it differ from Telegram trading bots?
Working principle: Perception, reasoning, and autonomous decision-making
AI Agent is an intelligent agent system based on large language models (LLM) that can perceive the environment, make reasoning decisions, and complete complex tasks through tool invocation or action execution. Workflow: Perception module (input acquisition) → LLM (understanding, reasoning, and planning) → Tool invocation (task execution) → Feedback and optimization (validation and adjustment).
Specifically, AI Agent first acquires data from the external environment (such as text, audio, images, etc.) through the perception module and converts it into structured information that can be processed. The LLM, as the core component, provides powerful natural language understanding and generation capabilities, serving as the system's 'brain.' Based on the input data and existing knowledge, the LLM conducts logical reasoning, generating possible solutions or formulating action plans. Subsequently, the AI Agent completes specific tasks by invoking external tools, plugins, or APIs, and validates and adjusts the results based on feedback, forming a closed-loop optimization.
In Web3 application scenarios, what is the difference between AI Agent and Telegram trading bots or automated scripts? Taking arbitrage as an example, users want to conduct arbitrage trades under the condition that profits exceed 1%. In a Telegram trading bot that supports arbitrage, the user sets a trading strategy with profits greater than 1%, and the bot starts executing. However, when the market is volatile and arbitrage opportunities change frequently, these bots lack risk assessment capabilities and execute arbitrage as long as the profit condition is met. In contrast, AI Agent can automatically adjust its strategy. For example, when a transaction's profit exceeds 1%, but data analysis assesses its risk as too high due to sudden market changes that could lead to losses, it would decide not to execute that arbitrage.
Thus, AI Agent has self-adaptability, and its core advantage lies in its ability to self-learn and make autonomous decisions. By interacting with the environment (such as the market, user behavior, etc.) and adjusting behavior strategies based on feedback signals, it continuously improves task execution effectiveness. It can also make real-time decisions based on external data and continuously optimize decision strategies through reinforcement learning.
Does this sound a bit like a solver under an intent framework? AI Agent itself is also a product based on intent, with the biggest difference from solvers under the intent framework being that solvers rely on precise algorithms, which have mathematical rigor, while AI Agent's decision-making relies on data training, often needing to approach the optimal solution through continuous trial and error during the training process.
Mainstream AI Agent frameworks
The AI Agent framework serves as the infrastructure for creating and managing intelligent agents. Currently, popular frameworks in Web3 include ai16z's Eliza, zerebro's ZerePy, and Virtuals' GAME.
Eliza is a multifunctional AI Agent framework built using TypeScript, supporting operation on multiple platforms (such as Discord, Twitter, Telegram, etc.) and capable of complex memory management, retaining previous conversations and contexts to maintain stable and consistent personality traits and knowledge responses. Eliza adopts a RAG (Retrieval Augmented Generation) system, allowing access to external databases or resources to generate more accurate answers. Additionally, Eliza integrates TEE plugins for deployment in TEE to ensure data security and privacy.
GAME is a framework that empowers and drives AI Agent to make autonomous decisions and actions. Developers can customize the agent's behavior according to their needs, expand its functionality, and provide customized operations (such as social media posting, replies, etc.). Different functions within the framework, such as the agent's environmental location and tasks, are divided into multiple modules for easy configuration and management by developers. The GAME framework divides the decision-making process of AI Agent into two levels: High-Level Planning (HLP) and Low-Level Planning (LLP), responsible for different tiers of tasks and decisions. High-Level Planning sets the overall goals and task planning for the agent, making decisions based on objectives, personality, background information, and environmental state, prioritizing tasks. Low-Level Planning focuses on execution, translating the decisions made in High-Level Planning into specific operational steps, selecting suitable functions and methods.
ZerePy is an open-source Python framework for deploying AI Agents on X. This framework integrates LLMs provided by OpenAI and Anthropic, enabling developers to build and manage social media agents that automate tasks such as posting tweets, replying, and liking. Each task can be assigned different weights based on its importance. ZerePy offers a simple command-line interface (CLI) for developers to quickly start and manage agents. Additionally, the framework provides Replit (an online code editing and execution platform) templates, allowing developers to quickly get started with ZerePy without complex local environment configuration.
Why does AI Agent face FUD?
AI Agent appears intelligent and can lower entry barriers and enhance user experiences. Why is there FUD in the community? The reason is that AI Agent is essentially still just a tool, and it cannot complete the entire workflow at present; it can only improve efficiency and save time at certain nodes. Moreover, at this stage of development, the role of AI Agent is mainly concentrated on helping users issue MeMe with one click and manage social media accounts. The community jokingly states, 'assets belong to Dev, liabilities belong to AI.'
However, just this week, aiPool released an AI Agent for token presale, utilizing TEE technology to achieve trustlessness. The wallet private key of this AI Agent is dynamically generated in a TEE environment to ensure security. Users can send funds (e.g., SOL) to the wallet controlled by the AI Agent, which then creates tokens based on set rules and launches liquidity pools on DEX, distributing tokens to eligible investors. The entire process is autonomously completed by the AI Agent in a TEE environment without relying on any third-party intermediaries, avoiding the common rug pull risks in DeFi. It is evident that AI Agent is gradually evolving. I believe that AI Agent can help users lower barriers and enhance experiences, even if it's just simplifying parts of the asset issuance process, it is meaningful. However, from a macro Web3 perspective, AI Agent, as an off-chain product, currently only serves as an auxiliary tool for smart contracts, so there is no need to overhype its capabilities. Due to the lack of significant wealth effect narratives besides MeMe in the second half of this year, it is normal for the hype around AI Agent to revolve around MeMe. Relying solely on MeMe cannot sustain long-term value, so if AI Agent can introduce more innovative gameplay in transaction processes and provide tangible landing value, it may develop into a common infra tool.
