什么是人工智能代理？它们与 Telegram 交易机器人有何不同？

自今年下半年以来，人工智能代理的话题一直备受关注。最初，真相的AI聊天机器人终端因其在X上的幽默帖子和回复（类似于微博上的“罗伯特”）而引起广泛关注，并获得了a16z创始人马克·安德森（Marc Andreessen）的5万美元资助。

AI Agents, a hot topic in the Web3 community, have sparked discussions and debates. But what exactly are AI Agents? How do they differ from Telegram trading bots? And why do they face skepticism despite their potential benefits?

AI Agents作为Web3社区的热门话题，引发了讨论和争论。但人工智能代理到底是什么？它们与 Telegram 交易机器人有何不同？尽管它们有潜在的好处，但为什么它们仍面临怀疑？

AI Agents are intelligent agent systems powered by large language models (LLMs) that can perceive their surroundings, make logical decisions, and complete complex tasks by utilizing tools or executing actions. Their workflow involves:

AI 代理是由大型语言模型 (LLM) 提供支持的智能代理系统，可以感知周围环境，做出逻辑决策，并通过利用工具或执行操作来完成复杂的任务。他们的工作流程包括：

* Perception module (acquiring input)

* 感知模块（获取输入）

* LLM (understanding, reasoning, and planning)

* LLM（理解、推理和规划）

* Tool invocation (task execution)

* 工具调用（任务执行）

* Feedback and optimization (validating and adjusting)

* 反馈和优化（验证和调整）

For example, in the context of Web3 applications, AI Agents differ from Telegram trading bots or automation scripts in the following way:

例如，在 Web3 应用程序的上下文中，AI 代理与 Telegram 交易机器人或自动化脚本的不同之处如下：

Suppose users want to execute arbitrage trades when profits exceed 1%. In a Telegram trading bot that supports arbitrage, users can set a trading strategy for profits greater than 1%, and the bot will begin executing trades that meet this condition. However, these bots lack the ability to assess risk and will continue executing arbitrage trades as long as the profit condition is met. In contrast, AI Agents can automatically adjust their strategies. For instance, if a trade's profit exceeds 1%, but data analysis reveals that the risk is too high due to potential sudden market changes that could lead to losses, the AI Agent will decide not to execute the arbitrage trade.

假设用户想要在利润超过1%时执行套利交易。在支持套利的 Telegram 交易机器人中，用户可以设置利润大于 1% 的交易策略，机器人将开始执行满足此条件的交易。然而，这些机器人缺乏评估风险的能力，只要满足盈利条件就会继续执行套利交易。相比之下，人工智能代理可以自动调整策略。例如，如果一笔交易的利润超过1%，但数据分析显示由于潜在的市场突然变化可能导致损失，风险过高，AI Agent将决定不执行套利交易。

Thus, AI Agents possess self-adaptability, with their core advantage being the ability to self-learn and make autonomous decisions. Through interaction with the environment (such as market conditions, user behavior, etc.), they adjust their behavioral strategies based on feedback signals, continuously improving the effectiveness of task execution. They can also make real-time decisions based on external data and continuously optimize decision-making strategies through reinforcement learning.

因此，AI Agent具有自适应能力，其核心优势是自我学习和自主决策的能力。他们通过与环境（如市场状况、用户行为等）的交互，根据反馈信号调整自己的行为策略，不断提高任务执行的有效性。他们还可以根据外部数据做出实时决策，并通过强化学习不断优化决策策略。

While AI Agents sound advanced and capable of enhancing user experiences, they also face skepticism in the community. This is mainly because AI Agents are still just tools and cannot complete entire workflows independently. They can only enhance efficiency and save time at certain nodes. Moreover, at the current stage of development, the role of AI Agents is mostly concentrated on helping users issue MeMes and manage social media accounts. As a result, the community戲稱" , poking fun at the fact that assets ultimately belong to the developer, while liabilities are assigned to the AI.

虽然人工智能代理听起来很先进并且能够增强用户体验，但它们也面临着社区的质疑。这主要是因为AI Agent仍然只是工具，无法独立完成整个工作流程。它们只能在某些节点提高效率并节省时间。而且，在目前的发展阶段，AI Agent的作用大多集中在帮助用户发布MeMe和管理社交媒体账户。于是，社区戏称“资产最终属于开发者，负债则归AI所有”。

However, just this week, a new application of AI Agents emerged with the launch of an AI Agent for token presale by aiPool. This AI Agent leverages TEE technology to achieve trustlessness. The wallet private key of this AI Agent is dynamically generated in a TEE environment, ensuring security. Users can send funds (such as SOL) to the wallet controlled by the AI Agent, which then creates tokens according to set rules and launches a liquidity pool on a DEX, while distributing tokens to eligible investors. The entire process does not rely on any third-party intermediaries and is fully completed autonomously by the AI Agent in a TEE environment, avoiding the common rug pull risks in DeFi. It is evident that AI Agents are gradually evolving. I believe that AI Agents can help users lower barriers and enhance experiences, and even simplifying part of the asset issuance process is meaningful. However, from a macro Web3 perspective, AI Agents, as off-chain products, currently serve merely as auxiliary tools for smart contracts, so there is no need to overstate their capabilities. Given that there has been a lack of significant wealth effect narratives aside from MeMe in the second half of this year, it is normal for the hype around AI Agents to revolve around MeMe. Relying solely on MeMe cannot sustain long-term value, so if AI Agents can bring more innovative gameplay to trading processes and provide tangible value, they may develop into a common infra tool.

然而就在本周，随着aiPool推出用于代币预售的AI Agent，AI Agent的新应用出现了。该AI Agent利用TEE技术实现去信任化。该AI Agent的钱包私钥是在TEE环境中动态生成的，保证了安全性。用户可以将资金（例如SOL）发送到AI代理控制的钱包，然后AI代理根据设定的规则创建代币，并在DEX上启动流动性池，同时将代币分配给符合条件的投资者。整个过程不依赖任何第三方中介，完全由AI Agent在TEE环境下自主完成，避免了DeFi中常见的拉扯风险。很明显，人工智能代理正在逐渐发展。我相信AI Agent可以帮助用户降低门槛、提升体验，甚至简化部分资产发行流程也是有意义的。但从宏观的Web3角度来看，AI Agent作为链下产品，目前仅作为智能合约的辅助工具，无需夸大其能力。鉴于今年下半年除了 MeMe 之外缺乏显着的财富效应叙事，围绕 AI Agent 的炒作围绕 MeMe 是很正常的。仅仅依靠 MeMe 无法维持长期价值，因此如果 AI Agent 能够为交易流程带来更多创新玩法并提供有形价值，它们可能会发展成为一种通用的基础设施工具。