"Humanity is what it is because it uses tools."
From an evolutionary perspective, the development of living organisms is mainly achieved through two methods: unit enhancement and organizational enhancement. These two enhancement methods complement each other, enabling life to possess more diverse and complex forms of expression. Just like an Agent—we hope it is an intelligent entity capable of independent thought and interaction with its environment in any system.
The success or failure of Agents will be key in determining whether this GPT revolution is a new generation of industrial revolution.
The term Agent originates from the Latin word Agere, meaning 'to do.' In the context of LLM, an Agent can be understood as a type of entity capable of autonomous understanding, planning decisions, and executing complex tasks.
An Agent is not an upgraded version of ChatGPT; it not only tells you 'how to do' but also helps you do it. If CoPilot is the co-pilot, then the Agent is the main driver. Similar to the human process of 'doing things', the core functions of an Agent can be summarized in a cycle of three steps: perception, planning, and action. This process is like the 'praxis' of Marxism: 'Understanding begins with practice, theoretical understanding is attained through practice, and returns to practice.' The Agent evolves in this unity of knowledge and action.
We can imagine the process of our interaction with the external environment: based on all our perceptions of the world, we deduce its hidden states and combine our memories and understanding of the world to make planning, decisions, and actions; while actions will also feedback into the environment, providing us with new feedback, which we observe to make further decisions, creating a continuous cycle.
Currently, Agents are still like cavemen. In the Generative Agents experiment of the simulation life game, each character is controlled by an AI Agent, living and interacting in a sandbox environment, fully demonstrating the process of transforming feedback and environmental information into actions, achieving the 'social' aspect of AI Agents.
Looking back at the common characteristics exhibited by the crypto market during several market cycles we have experienced. Like DeFi, NFTs, or the 'metaverse', each market cycle creates speculative markets and also generates some inflated technological imagination. Overheated speculative markets not only drive liquidity inflows but also satisfy high-quality labor and abundant capital, accelerating technology adoption. After the short-term inflation of market interest subsides, participants with fundamental value remain in the market, maturing the industry and transcending short-term narratives.
If we believe that cryptocurrencies and AI agents have real potential, rather than just being a fleeting narrative in this market cycle, we need to discuss the compatibility of cryptocurrencies and AI agents from a longer-term perspective.
Looking back at previous examples, when non-blockchain native technologies or industries combine with cryptocurrencies, they typically develop within a structure that benefits both parties. For instance, the integration of traditional finance and DeFi is a prime example. Traditional financial infrastructure can create flexible primary and secondary markets through DeFi. Conversely, DeFi diversifies collateral types through traditional assets such as U.S. Treasury bonds, ensuring a stable collateral structure. Similarly, positive mutual influences can also arise when other technologies or industries combine with cryptocurrencies.
The payments market has clearly demonstrated that payment channels not restricted by traditional financial infrastructure or national borders are one of the greatest value propositions of cryptocurrencies. Likewise, combined with AI agents, cryptocurrency payment channels provide effective solutions for enhancing the performance of AI models.
At the same time, cryptocurrencies can explore various developmental possibilities through AI agents. In particular, the 24/7 operations of blockchain and cryptocurrency markets require operators to work around the clock. In this regard, similar to the basic functions of AI agents, autonomous agents have the potential to streamline most on-chain interactions.
Most AI agents can simplify interactions within cryptocurrencies. For example, Griffain @griffaindotcom autonomously executes on-chain interactions based on user prompts, while Zerebro @0xzerebro has proposed a development plan for AI agents that autonomously execute Ethereum network validator operations.
While these are just simple examples, there is sufficient synergy between cryptocurrencies and AI agents across a wide range of fields like security, on-chain user experience, privacy, or asset tokenization. Of course, these ideas are still in their infancy, and concepts like executing validator operations require meticulously designed technological cores.
After the overheating of the AI agent market subsides, clues will emerge to discern what will remain and what will not. Projects that can reasonably answer the question 'Why cryptocurrency?' will be Virtuals Protocol @virtuals_io and ai16z @ai16zdao. They are at the forefront of providing these answers, and many subsequent agents are attempting cryptocurrency integration in various ways. Additionally, multi-agent, intent-based interfaces, and alternative frameworks are driving the development of experimental environments.
As a16z's Dixon said: 'The next big thing initially looks like a toy.'
AI agents have evolved from merely generating response text on Twitter to being capable of executing complex tasks such as validators, white hat operations, and autonomous on-chain transactions.
Let's see together whether there will still be meaningful innovations at the end of this AI agent cycle or if it will simply become another forgotten hype cycle.
#ai16z #AIAgents #Crypto #特朗普上台概念币有哪些？
