Author: Ice Frog Source: X, @Ice_Frog666666

AI Agent is popular in the encryption world, and It seems that it is going to explode, and in the face of the huge wealth effect, it has been shaped into a new revolutionary variable, hot and unparalleled in the limelight. However, the word revolutionary is a highly inflationary concept in the crypto world. It is scattered in large and small white papers and social media. For Lao Lee, similar words are already difficult to make people passionate. There have been countless tokens in the crypto world. The rise and fall of Cryptozoology proves that only a handful of new narratives have made the crypto market rise, and the vast majority are just short-lived.

So, we must further explore a classic question in the currency circle: xxx, will it be different this time? Everyone has a different answer, but if you look back on the past, at least the fundamental conclusion of this question will not change much.

That is: the encryption world follows the attention economy. How far the new narrative can go depends on the extent to which user attention and network effects can be spread. Bitcoin cannot escape this fundamental law.

Naturally, if we want to analyze this problem, we must start from the source and explore the answer to the problem.

AI Agent refers to Artificial Intelligence Agent. Currently, the concept that is relatively well-known and commonly used by the public is: AI Agent (Artificial Intelligence Agent) is an agent that can perceive environment, intelligent entities that make decisions and perform actions. It is mainly based on LLM (Large Language Model), in other words, it is the functional carrier of large language model. The interesting point is that intuitively, it is a specific application of LLM, but the name is indeed Agent, which emphasizes being given the power and corresponding capabilities to make independent choices, actions, and decisions.

From the definition point of view, although the AI ​​Agent and the large model have a sequential relationship, the interaction between the large model and humans requires the help of prompts (prompt words) ) gives specific tasks and then answers. AI Agent is a human being who gives specific goals. The AI ​​Agent will dismantle the execution steps according to your goals, give itself prompts, and then achieve the completion of the goals.

From the perspective of collaboration between humans and AI, Agent mode is also a relatively advanced collaboration method. By analogy, it is the early Siri— Microsoft Coplilot—AI Agent. From this collaboration point of view, we can basically understand that AI Agent is essentially a digital mapping of human thinking and behavior patterns, so its structure is mainly a question and answer portal + fully automated workflow. (perception, decision-making, action) + Knowledge base (human hippocampus), AI does most of the work, not just assisting.

In terms of the specific technical framework, Lilian Weng, the former chief security researcher of Open AI, wrote a blog in June 2023 to elaborate specifically, titled "LLM Powered Autonomous Agents", as follows:

This article, the basic framework of AI Agent proposed by Lilian Weng = LLM + planning + memory + tool use, big The main function of the model is to assume functions similar to human brain reasoning and planning.

In general, AI Agent. It is a separate computing entity that is based on the reasoning and planning functions of a large model, combined with the perception of the external environment, the use of tools, and actions, so as to achieve the role of AI as a human agent and complete a series of relatively complex tasks.

From 2023, AI Agent will enter the industry's vision, and the discussion and progress of Agent will continue to accelerate. Most major manufacturers regard 2025 as the year AI Agent In a critical year for the commercial explosion, the industry is entering a period of accelerated development.

From the overall perspective of the industry chain, the upstream of the entire industry chain is still dominated by Nvidia and others. The main ones are computing power and hardware suppliers, data suppliers, algorithm and large model development and other infrastructure providers. The midstream is mainly AI Agent integrators, and the downstream is vertical applications or general agents for different industries.

The current Ai Agent industry, regardless of the existing infrastructure in the upstream, is mainly concentrated in the midstream and downstream, especially the applications in the downstream. A hundred flowers are blooming.

In addition, from the perspective of downstream development, the C-side and The B-side has different progress, and the C-side uses AI Agent It can greatly improve the user experience, and the use of AI Agent on the B-side can significantly reduce costs and increase efficiency.

Judging from the actions of several major manufacturers. , the current AI Agent has begun to accelerate in the second half of this year, and is expected to accelerate overall next year. When Google released Gemini 2.0 this month, it not only emphasized that the model is mainly. To serve AI Agent, Project was released Astra (general purpose), Project Mariner (browser operation) and Jules (programming) three AI Agent products

At the end of October this year, Microsoft launched Dynamics 365. 10 AI Agents have been launched on the platform

Amazon announced this month that it will open an artificial intelligence laboratory in San Francisco to focus on the implementation of Ai Agent

Open AI has launched 12 consecutive days of new product releases this month. Sam Altman himself also declared that next year will be the year when AI Agents enter the mainstream. Although Open AI itself has not released related AI agents, in terms of basic framework A series of tools have been launched to support the development of AI agents.

Whether it is a large manufacturer or from the perspective of industry prosperity, the current AI Agent has indeed entered the market. After an acceleration period, it entered Crypto It's a matter of time, but everything is still in its early stages.

Turn the clock back to a few months ago. , I'm afraid the And that created the Truth of Terminal modely Ayrey never expected that an Ai Agent model, which was still an experimental product, would actually create a shocking wealth miracle in the crypto world. The GOATSE concept derived from it became a hot spot in crypto. The MEME token GOAT became popular in half a day. It soared to 20 million U.S. dollars, with a market value of nearly 300 million U.S. dollars within four days, and a market value of over 1 billion U.S. dollars within one month. The miracle of a thousand-fold increase reappeared. AI Agent entered the crypto world in a very Crypto way and caused a huge wave of enthusiasm.

Subsequently, Ai16Z (Ai Agent-driven venture capital fund) quickly became popular with the concept of "AI-driven DAO", coupled with the founder of orthodox A16z Marc Andreessen Supported, the increase exceeded 10 times in a few days; after that, another AI project ACT was listed on Binance, and the AI ​​Agent was directly It has reached a new climax, not only has the market volume been unprecedented, but it has also allowed the majority of crypto users to deeply understand the wealth effect of this track.

In the first two chapters of this article, a lot of space is spent to clarify what exactly is AI Agent and the industrial development of AI Agent. The important significance is that the source of AI is not Coming from the world of encryption, in a broader sense, encryption is not the main battlefield for AI, but if AI Agent does not develop well in a broad sense, attention to AI Agent will quickly dissipate, and the narrative of AI Agent will become short-lived like most narratives.

However, this time, as far as the naked eye can see, it is indeed different. The main reason is:

The AI ​​world in the broad sense has not entered the bubble falsification period. Whether it is Nvidia, Microsoft, Google and other major manufacturers, in their third quarter reports this year The top four manufacturers alone will invest more than US$170 billion in capital next year, with only one goal: AI.

From the perspective of the development of AI Agent, although there is currently no phenomenal product that has exploded the market like chatgpt, the actions of major manufacturers and the industry In terms of development, its booming momentum and financial bets are rising sharply. There is a certain probability that an AI Agent that will detonate the market will appear in 2025.

From these two points, from a broad scientific and technological perspectiveOn the Internet, the topic and attention of AI and AI Agent are still the hottest topics in the market. As mentioned at the beginning, attention in the encryption world is everything.

From the perspective of the implementation technology framework of AI Agent, its combination with Crypto will give birth to a key turning breakthrough similar to the birth of Ethereum smart contracts. Not only It is the addition of technology, and it is more likely to lead to a leap-forward economic paradigm change, because it will essentially change the creation model of the attention economy.

The biggest problem that blockchain currently affects Mass Adoption may be its complex operations and entry thresholds. Putting aside the compliance pain points involving legal currency such as deposits/withdrawals, The difficulty and complexity of on-chain operations and wallet operations are at least several times that of Web2. If you use the natural language mode of Ai Agent, a simple instruction can not only greatly simplify the operation difficulty, whether it is wallet management, screening of the best Defi investments, cross-chain, automatic execution of trading plans based on external market conditions, etc. It directly reduces the learning cost of new users by several levels.

In addition, whether it is the creator economy, market sentiment monitoring, smart contract auditing, governance voting, AI autonomous DA or even MEME issuance, you can use Agent to participate. Under a given set of circumstances, it may be more conscientious and fair than most people, and more able to reject the influence of emotions.

From a narrative logic, what AI+Crypto brings is: AI can make trusted blockchains smarter, and blockchains can make intelligent AI More credible.

The characteristic of blockchain is the non-tamperability of data. A major drawback of AI happens to be data quality. If the built AI Agent can train the data on the chain and Using its computing power, it is likely to change the current incentive model.

In the longer term, perhaps in the near future, every crypto user will have a digital clone, which will help you manage your token assets. Social interaction, etc.; each project party has several AI Agents to help operations from asset issuance, marketing, code construction, contract auditing, media operations, and even airdrop design and distribution, etc. can be completed with the help of AI.

These long-term changes will change the creation model of attention, from larger communities and humans to AIAgent.

Of course, the above ideas are only long-term fantasies. Returning to reality, the current AI Agent is still in an early barbaric era in the blockchain, except for AI In addition to the popularity of MEME, AI Agent is still in the stage where speculation is greater than build. There is currently no AI Agent framework based on blockchain features on the market. Even the first ELIZA is only at the conversation level. It has not yet entered the core of the entire blockchain world.

From the three levels of AI Agent, the perception, decision-making, and execution levels all need to rely on decentralized features and smart contract features to build a more systematic and underlying system. Infrastructure, not to mention large and small tool platforms, data privacy, transaction security, etc. The good news is that whether it is the attention of top funds such as A16z to AI Agent, the mainstream narrative of AI Agent in the world in a broad sense, or the amazing wealth effect that AI MEME has exploded, AI Agent has a solid foundation for development in the crypto world. Base.

In the encrypted attention economy, the current AI Agent has the attention of capital, the continued popularity of broad narratives, typical cases of wealth effects, and long-term Practical value of construction.

Perhaps, we can be a little bolder and say that this time, it is really different!

Passion!