Author: Haotian
In my previous article, I shared the logical relationship between AI DePIN, AI MEME, and AI Agent narratives, but when it comes to exploring the potential Alpha of AI, the problem arises - is it the AI MEME with strong wealth effects but frequent replacement from the conspiracy group? Is it the early-stage protocol framework that is still building the infra? Or should we choose some cool and specific application scenarios of monolithic AI? To be honest, I've been confused for a long time, and here are some of my shallow insights:
1) The overall trend is undoubtedly that the AI Agent is not a "asset" or an "application", but a new paradigm of large-scale blockchain ecosystem "narrative".
In other words, the arrival of the AI Agent will be like the DeFi Summer of 2020 in the blockchain industry, and it will spawn a huge new "bubble" narrative cycle around the Token issuance, custody, and trading of the AI Agent; the construction of infrastructure, frameworks, standards, communication, stacking, and SaaS encapsulation; the development, operation, aggregation, iteration, and implementation of applications.
DeFi has driven the glory of the last bull market, and the three pillars of stablecoins + lending + DeX derivatives have ignited tremendous market vitality. Incomplete statistics show that the total locked value of multi-chain DeFi has reached 134.6 billion, the circulation of stablecoins has exceeded 100 billion US dollars, the monthly trading volume of DEX has reached 372.3 billion US dollars, and the lending volume is close to 20 billion US dollars. The market size and ecological importance of the entire DeFi industry are growing at an astonishing rate and are still continuously refreshing, not to mention the broader data such as MCAP\FDV of various chains and applications generated by DeFi.
In my view, the AI Agent will also create a new narrative miracle that may even surpass the DeFi narrative, with an imagination space of 200B, 500B. Stop always thinking about a new DeFi Summer 2.0, please keep up with the current AI Agent Summer narrative tide, don't fall behind.
2) Based on this logic, let's take a look at the "chaos" in the current AI Agent market:
1) The community battles and the competition for the top position among AI MEME projects like $GOAT, $FARTCOIN, and $ACT; 2) The competition for AI Agent framework standards like $VIRTUAL and $AI16Z; 3) The competition for derivative infra like asset issuance and circulation such as $VVAIFU and $GRIFFAIN; 4) The competition for monolithic AI Agent applications like $AIXBT, $AVA, and $BULLY, etc.
You will find that it is still too early. At least the various lending platforms, derivatives, yield farming, and governance tokens that appeared in DeFi had the basic execution framework and standards provided by Ethereum and other chains, while the AI Agent ecosystem doesn't even have dedicated infra conditions yet. From the perspective of ecosystem plasticity, the industry narrative opportunities brought by AI Agent are much greater than DeFi.
In other words, this chaos is not chaotic. If you are optimistic about the development of the AI Agent narrative, for each sector, the assets that have reached a certain market value and have a corresponding user base are quality assets before the massive AI Agent narrative explosion, just like the assets like YFI, CRV, UNI, and DYDX that have reached hundreds of billions of FDV in the past.
Of course, confidence does not mean you can mess around. Don't forget that in addition to the "big aunt", there are also "second aunts" like pseudo-value assets, and these assets always quietly invade your wallet with a more tempting but long-term toxic attitude.
3) Because it is indeed still in the early stage of the narrative, there is no reason to persuade everyone not to participate in the AI MEME that is clearly manipulated by the conspiracy group. After all, no one knows how much wealth effect an early track can unleash, and the frameworks and standards that seem more valuable and have been implemented may also face the possibility of being surpassed by competitors at any time.
In short, it is difficult to give a clear Alpha exploration logic. I can only share my personal research logic for reference:
1. For AI MEME, only consider the "leader coins" like $GOAT that have the nature of the narrative pioneer, without logic. If you have to say one, it's just like Bitcoin, the consensus coin, gave birth to Ethereum, the smart contract technology delivery coin.
An early MEME in a track, although the speculative attribute is the main one, its role in educating the market and cultivating user cognition cannot be ignored, just like DOGE's popularization of cryptocurrencies in the early days.
2. AI Agent infra is very important, and it is undervalued in the long run. For example, the early frameworks like $ELIZA with rich Github open-source libraries and wide application coverage, although they are indeed very early and may be drained by other stronger framework standards at any time, their absolute advantages of open-source sharing and first-mover advantages are the barriers. The frameworks that are even better than it will most likely be on top of it, which may affect its growth height, but they will definitely not eliminate it.
We treat ELIZA as the EVM of the AI Agent era, and there will certainly be new frameworks like EVM++ and EVM Compatible to supplement it.
3. For AI Agent asset issuance, custody, and trading, it is currently still at the MEME-ized asset issuance narrative level. We need to lock in the platform value instead of being confused by the "junk assets" on the platform.
The logic is simple. The essence of AI Agent is the process of AI scenario value landing based on Crypto incentive mechanisms. The early infra, applications, protocol frameworks, etc. are not yet mature, how could there be so many AIXBT-level super applications? What can the convenience of token issuance bring? Understand it? In the end, it's just a bubble trap to create heat for assets like VIRTUAL and VVAIFU.
Therefore, don't chase AI Agent super applications too much, unless you have absolute information advantages, otherwise, by the time you see the market value rushing in, the cost performance will not be high, but locking in innovative platform infra is different.
4. There is still a vast blank space in the construction of AI Agent infra. For example, the early frameworks like ELIZA have only solved the problems of LLM large models and AI Agent dialogue mechanisms, and have been applied to scenarios like Twitter and Discord, but they are still mainly the matching of Web2 information. When will a version of ELIZA++ come out that can connect the ELIZA framework and Web3 trading scenarios based on the public chain paradigm?
Also, platforms like Virtual and VVAIFU can quickly issue assets, but they still lack a lot in the direction of decentralized asset custody, AMM-style intentional trading, etc. When will the AI Agent+TEE private key management method discussed earlier mature? Can the asset issuance platform extend new gameplay in trading, and come up with more interesting trading gameplay and incentive models?
Furthermore, can there be a chain that builds interaction standards for AI Agents, becoming the Cosmos of the AI Agent era, providing standard interoperability capabilities? Or some chains that build modular storage for the permanent Memory of AI Agents, providing the necessary DA capabilities for AI Agents? There are also some AI Agent visualization tools, execution engines, modular combination encapsulation services, etc.
With these technically and market-driven supporting infra coming out, will you still be anxiously guarding a bunch of MEME that you don't understand?
As for where the previous infrastructure built based on the smart contract paradigm, such as layer1, high-performance chains, cross-chain infra, chain abstraction, DA composite chains, layer2, reStaking, etc. should go, in addition to the original infra landing application PMF delivery problem, a breakthrough point is the new concept of "riding the AI Agent narrative".
So, which chain can stand out to provide some communication protocols, DA, application development and other service capabilities for the AI Agent, they are all worth paying attention to. Of course, there must be some chains that are purely riding the concept, but they are still better than the chains that are stuck in the old ways.
In any case, please lock in the AI Agent track and welcome the arrival of this epic bull market main wave.
