Author: Haotian
Perhaps everyone feels that this round of AI Agent narrative progression is very similar to the inscription track since 2023. While not entirely equivalent, this comparison makes a lot of sense. I believe we can at least replicate the historical experiences of the inscription track to the AI Agent track, sorting out some potential development evolution logic and investment opportunities:
- The beginning of the bubble in the 'asset issuance' narrative.
Undoubtedly, whether in inscriptions, AI Agents, or the NFT tracks of the past, the evolution of narratives has been birthed from the FOMO emotions of 'asset issuance.' After all, it was too early, and no one could fathom the future utility of such rapidly circulating tokens in the market. Perhaps it is precisely because of this information asymmetry that a potential 'get rich quick effect' arises. This will drive the FOMO sentiment to spread like a snowball, causing more people to search for wealth secrets, resulting in a 'bubbling' situation where more meaningless tokens are born. This is quite difficult to avoid and rather serves as the engine for the emergence of successive bull markets.
For example, the initial hype of the inscription track revolved around BRC20, where the core selling point, aside from being novel and fun, was the scarcity of 'first is first,' focusing on the narrative advantage of being the pioneer of dragon one and dragon two. AI Agents are no exception; they initially hyped up AI interactive dialogues, with similar MEME attributes found in coins like Goat, act, and fartcoin. The script is also similar, with a prolonged MEME battle around dragon one and dragon two, and whether it has settled remains to be seen.
Of course, the narrative of AI Agent asset issuance is much stronger than that of inscriptions, as it at least has some utility application attributes. It is merely a question of how to enrich application scenarios like dialogue interaction, image generation, and on-chain transactions, and whether a single application scenario can support a market cap of hundreds of millions.
But I want to remind you that AI Agents inherently possess application attributes; therefore, the process of the maturity of application tracks is also a continuous process of stripping away MEME-ization. If you must choose a MEME, find the one that best represents the cultural attribution of AI Agents.
- The competition for 'framework standards' is a grand spectacle of internal competition.
When the MEME-driven pure financial nihilism 'asset issuance' narrative reaches a certain stage, it will inevitably transform into a narrative of technological empowerment, launching a longer cycle of competition centered around the differences in technical standards and framework protocols.
Looking back at the emergence of ARC20, SRC20, DRC20, Runes protocols, and other public chain inscription pressure test battles, doesn't it resemble the current chaotic battle among frameworks and standards like ELIZA, ARC, Virtual, Vvaifu, zerobro, Griffain, and Swarms?
Throughout this process, the hotspots of the track will continue to be refreshed, and the same funds will continuously rotate. Communities around technology will also emerge, and there will certainly be some conflicts and verbal battles.
The so-called existence of a technical melee is primarily because, at this stage, it is difficult for technology to prove itself superior or inferior. Using factors like the number of GitHub repo forks, star counts, the professional background of initiators, and mysterious investors as reference points for judgment cannot withstand the huge fluctuations of a project. Because you never know which piece of information might be misleading.
However, since it is a competition of technical quality standards, the quality of technology must occupy the 'first place.' My process of sorting out and analyzing new project technical frameworks, features, advantages, etc., is also a kind of 'technical screening.' Although it may not be accurate, at least it can avoid being misled by some MEME coins disguised in technical clothing. The logic is simple: the threshold for technical project fraud is not low. The challenge lies in the aesthetic preferences of the entire developer community; one cannot attempt to harvest based on a flashy website alone.
In general, the market manifestation at this stage is 'hotspot rotation,' mixed with various fraudulent projects. Fortunately, projects that can gain attention by selling technical quality are generally not bad. Of course, do not feel FOMO for missing out on a new project; FUD will definitely exist (it’s the fate of rapid market cycles), and opportunities to enter during significant fluctuations will certainly arise. Even if you miss out, you must believe that there will be a better version than it. During this stage, adopting a playboy mentality in cryptocurrency is certainly not wrong.
There is no way out; even the most perfect framework standard design starts as just an idea, akin to financing during the ICO era relying solely on white papers. The best logic is that the competition lies in the digestion of information and learning comprehension. I believe the market is still innovating and iterating, and there will always be better technical frameworks emerging.
As long as you select a project with good quality, continuous iterative progress, and a cost-effective market cap to align with, you can simply wait for time to reward you.
- The integration phase of AI Agent 'chainization.'
Building around the technical standard framework of AI Agents is a prerequisite for the 'chainization' of AI Agents. A key issue is that frameworks like ELIZA, ARC, and Swarms are designed for the rapid deployment of AI Agents and to build powerful AI Agent capabilities, thereby allowing AI Agents to better serve humanity. Following this logic, it becomes clear that AI Agents in the web3 domain cannot be limited to graphic and text information interaction. Is it merely for entertainment?
If AI Agents want to lock in native crypto characteristics, they must embrace 'transaction characteristics.' How to fully combine the automated indexing of knowledge, information processing, optimized decision-making capabilities of AI Agents with the automation and composable invocation characteristics of on-chain smart contracts is the vision that AI Agents must achieve to connect with blockchain.
Moreover, the issues of memory storage in AI Agents, multi-modal interaction resource optimization, and trustworthy interaction also need to rely on the public, transparent, and decentralized characteristics of blockchain to be resolved. This does not even include the benefits brought about by the incentive mechanisms of Tokenomics. Therefore, it can be asserted that AI Agents will definitely 'chainize.'
Looking back after several waves of narrative in the issuance of inscription assets, the market ultimately arrived at VC and developers successively laying out value applications on the BTC layer 2 infrastructure narrative.
AI Agents will also follow this path. After experiencing a long period of PVP in asset issuance and a chaotic battle over framework standards, a large number of old infrastructure chains will embrace AI Agents, and some infrastructure will emerge to provide chainization services for AI Agents. At that time, decentralized AI will truly pump and become a major track at the level of DeFi Summer.
As small investors, it is certainly frightening. Are we to let retail investors become cannon fodder in a bubble narrative, leaving the fruits to VCs? Not at all, because the MEME-first approach of AI Agents, building community paths, and the past method of VCs injecting funds into projects in phases and finally delivering market expectations before TGE are fundamentally different.
VCs will definitely enter with great fanfare, but they may change the game. Perhaps the model of MEME coins leading the way and then subsequently empowering technical applications will become a new norm in Tokenomics.