Background: Crypto + AI, looking for PMF
PMF(Product Market Fit) refers to the product market fit, which means that the product should meet the market demand, confirm the market situation before starting a business, understand what type of customers to sell to, and understand the market environment of the current track and then develop the product.
The PMF concept applies to entrepreneurs to avoid creating a product/service that they feel good about but the market won't pay for, and the same concept applies to the cryptocurrency market, where the project should understand the needs of the players in the coin community to build a product, rather than building a technology that is out of step with the market.
In the past, Crypto AI was mostly bundled with DePIN, and the narrative was to use the decentralized data of Crypto to train AI, so as to avoid relying on the control of a single entity, such as computing power, data, etc., and the data provider could share the benefits brought by AI.
According to the above logic, in fact, it is more like Crypto enabling AI, in addition to distributing the benefit of the token to the computing power provider, it is difficult to Onboard more new users, it can be said that the model is not so successful on the PMF.
The emergence of AI Agent is more like the application side, compared with DePIN + AI like infrastructure, and obviously the application is relatively simple and easy to understand, but also has a better ability to absorb users, and has a better PMF than DePIN + AI
Sponsored by Marc Andreessen, founder of A16Z (PMF theory was also proposed by him), GOAT, which was generated by two AI dialogues, opened the first shot of AI Agent, and now ai16z and Virtual have their own advantages and disadvantages. What is the development track of AI Agent in the coin circle? What stage is it at? Where does the future go from here? Let WOO X Research take a look.
Stage one: The meme starts
Before GOAT, the most popular track this cycle was memecoin, which is characterized by inclusiveness, from the zoo's hippo MOODENG, to the DOGE owner's newly adopted Neiro, the Internet native meme Popcat, etc., showing the trend of "everything can be memed". And this seemingly absurd narrative, in fact, also provides the soil for AI Agent growth.
GOAT is a meme-coin created by a dialogue between two AI's, the first time AI has used cryptocurrencies and networks to achieve its goal of learning from human behavior. Only Meme coins can carry such a high experimental nature of the project, at the same time, similar concept coins have sprung up, but most of the functions stay in the automatic tweet, reply, etc., no practical application, at this time AI Agent currency is usually called its AI + Meme.
Representative project:
Fartcoin: Market cap 812M, on-chain liquidity 15.9M
GOAT: Market cap 430M, on-chain liquidity 8.1M
Bully: The market value is 43M, and the on-chain liquidity is 2M
Shoggoth: Market cap 38M, on-chain liquidity 1.8M
The second stage: Explore the application
Gradually, people are realizing that AI agents can not only have simple interactions on Twitter, but can be extended to more valuable scenarios. This includes content production such as music videos, but also investment analysis, money management and other services that are more suitable for users of the coin circle, from this stage on, AI agents and memes are separated, thus forming a new track.
Representative project:
ai16z: Market value 1.67B, on-chain liquidity 14.7M
Zerebro: Market cap 453M, on-chain liquidity 14M
AIXBT: Market value 500M, on-chain liquidity 19.2M
GRIFFAIN: Market cap 243M, on-chain liquidity 7.5M
ALCH: Market cap 68M, on-chain liquidity 2.8M
Introduction: Distribution platform
When AI Agent applications bloom, what kind of track can entrepreneurs choose to grasp this wave of AI and Crypto?
The answer is Launchpad
When the currency under the issuance platform has a wealth effect, users will continue to find and buy tokens issued by the platform, and the real income generated by the purchase of users will also enable the Taiwan dollar to drive up the price, and the Taiwan dollar price continues to rise, funds will spill over to the currency issued under the platform, forming a wealth effect.
The business model is clear and has a positive flywheel effect, but there are still points to note: Launchpad belongs to the winner-take-all Matthew effect, Launchpad's core function is to issue new tokens, in the case of similar functions, the competition is the quality of its projects, if a single platform can produce high-quality projects stably, and has a wealth creation effect, users will naturally increase the degree of adhesion to the issuance platform. And it's hard for other projects to steal users.
Representative project:
VIRTUAL: Market capitalization 3.4B, on-chain liquidity 52M
CLANKER: Market cap 62M, on-chain liquidity 1.2M
VVAIFU: Market cap 81M, on-chain liquidity 3.5M
VAPOR: Market value 105M
Stage 3: Seek collaboration
After the AI Agent began to achieve more practical functions, it began to explore collaboration between projects to build a more robust ecosystem. The focus of this phase is on interoperability and the expansion of the ecological network, in particular whether synergies can be created with other cryptographic projects or protocols. For example, AI Agents may work with DeFi protocols to enhance automation investment strategies, or integrate with NFT projects to enable smarter tools.
To achieve effective collaboration, you first need to establish a standardized framework that provides developers with preset components, abstractions, and related tools to simplify the development process of complex AI agents. By providing standardized solutions to common challenges in AI Agent development, these frameworks help developers focus on the uniqueness of their respective applications rather than designing the infrastructure from scratch each time, thereby avoiding the problem of reinventing the wheel.
Representative project:
ELIZA: Market cap 100M, liquidity 3.6M on the chain
GAME: Market cap 237M, on-chain liquidity 31M
ARC: Market value 300M, on-chain liquidity 5M
FXN: Market cap 76M, on-chain liquidity 1.5M
SWARMS: Market cap 63M, liquidity on the chain 20M
The fourth stage: Fund management
At the product level, AI agents may play more of a simple tool role, such as giving investment advice and generating reports. However, fund management requires a higher level of competence, including strategy design, dynamic adjustment, and market prediction, which signals that AI agents are not just tools, but are beginning to participate in the process of value creation.
As traditional financial funds accelerate their entry into the crypto market, the demand for specialization and scale continues to increase. The automation and efficiency of AI agents can complement this need, especially when performing functions such as arbitrage strategies, asset rebalancing and risk hedging, which can significantly improve the competitiveness of funds.
Representative project:
ai16z: Market value 1.67B, on-chain liquidity 14.7M
Vader: Market cap 91M, on-chain liquidity 3.7M
SEKOIA: Market cap 33M, on-chain liquidity 1.5M
AiSTR: Market cap 13.7M, on-chain liquidity 675K
Expectation Stage 5: Reinventing Agentnomics
At present, we are in the fourth stage. Regardless of the currency price, most Crypto AI Agent has not been implemented in our daily life applications. Take the author as an example, the most frequently used AI Agent is Web 2's Perplexity, I occasionally read AIXBT's analysis tweets. In addition, the frequency of use of Crypto AI Agent is extremely low, so it may stay for a long time in the fourth stage and is not yet mature at the product level.
The author believes that in the fifth stage, AI Agent is not only the aggregation of functions or applications, but the reshaping of Agentnomics (Agent economics), the core of the entire economic model. This stage of development will not only involve technological evolution, but more importantly will redefine the token economic relationship between Distributor, Platform and Agent Vendor, creating a new ecosystem. The following are the main features of this stage:
1. Compare the development history of the Internet
The formation of Agentnomics can be likened to the evolution of the Internet economy, such as the birth of super applications such as wechat and Alipay. By integrating the platform economy, these applications bring standalone applications into their ecosystem and become multi-functional portals. In this process, an economic model of collaboration and symbiosis was formed between application vendors and platforms, and AI Agents will repeat a similar process in the fifth phase, but based on cryptocurrency and decentralized technologies.
2. Reshape the relationship between distributors, platforms and Agent suppliers
In the AI Agent ecosystem, the three will establish a closely linked economic network:
Distributor: Distributor is responsible for marketing the AI Agent to end users, such as through professional app marketplaces or DApp ecosystems.
Platform: Provides the infrastructure and collaboration framework that allows multiple Agent vendors to operate in a unified environment, and is responsible for managing the rules and resource allocation of the ecosystem.
Agent Vendor: Develop and provide AI agents with different functions to deliver innovative applications and services to the ecosystem.
Through the design of the token economy, benefits between distributors, platforms, and vendors will be decentralized, such as sharing mechanisms, contribution returns, and governance rights, thereby facilitating collaboration and incentivizing innovation.
3. Entry and integration of super applications
When AI agents evolve into super application portals, they will be able to integrate multiple platform economies and absorb and manage a large number of independent agents. This is similar to how wechat and Alipay integrate independent applications into their ecology, and the super application of AI Agent will further break the traditional application island.