Original｜Odaily Planet Daily ( @OdailyChina )
Author: Wenser ( @wenser 2010 )
After the previous AI Agent concept coin hype, the Web3 AI project is in a rare cooling-off period. In view of this, I went to Hangzhou to participate in two AI-related activities with my curiosity about the Web2 AI project. One was an AI Hackathon with different participants and dazzling directions; the other was an offline community activity that used AI as a money-making tool. Here, I seem to have discovered some major differences between Web2 AI projects and Web3 AI projects, so I wrote this article. The following content is only personal subjective thinking and opinions, and does not represent the official opinion of Odaily Planet Daily. It is a small incision profile of the AI era for readers reference. (All dedicated AI projects are also welcome to join the construction and give a chance to witness the build.)
The biggest difference between Web2 AI and Web3 AI: one is making products, the other is making assets
In my opinion, there are many differences between Web2 AI projects and Web3 AI projects, but the most important difference is the result of precipitation - the former mainly speaks with products, whether it is a large model, AI application or AI solution; the latter is packaged with AI, and its essence is to create conceptual assets and judge heroes by the token market performance. This is also the reason why AI Agent concept tokens such as GOAT, AI16Z, ACT, swarms, etc. were highly sought after due to the AI Agent concept, but gradually fell into decline due to the shift of market attention. The following is an analysis of the differences between Web2 Web3 AI projects from a personal perspective——
Developer groups: Everyone is a Dev VS Technical Dev
This is my biggest feeling after participating in two Web2 AI offline activities: Web2 AI activities often cover a wider range of people, from children as young as eight or nine years old to grandfathers in their forties, all of whom have great enthusiasm for Web2 AI; while Web3 AI projects are often limited to the technical Dev group, and others join in more through token trading, project investment, etc. Although many AI Agent projects also focus on the concept of everyone can create their own AI Agent, there are few actual participants and it does not involve much development work.
The reason is that the entry threshold and narrow usage scenarios of Web3 have discouraged a large number of people; Web2 AI projects are closer to the Internet, so there are more, broader and more comprehensive developers, especially after the emergence of AI programming applications such as Cursor and Windsurf, it can be said that everyone is a Dev.
Two of the youngest contestants at the AI Hackathon
Project starting point: starting from demand vs starting from the market
As for the starting point of the project, Web2 AI projects often start from the needs of users, which is to solve a specific problem, create products in exchange for profits; while Web3 AI projects often start from the market. What kind of narrative, concept, and asset does the market need? Then Web3 AI projects focus on these aspects and seek financing. Due to the above reasons, generally speaking, Web2 AI projects focus more on the application layer; while Web3 AI projects tend to adopt the three-part method of computing power, algorithm, and data for project preparation, such as the previously popular Myshell, and the recently highly concerned projects such as Nillion and SaharaAI.
In comparison, the problems currently solved by mainstream Web3 AI projects may be - how to create a token and how to sell an AI concept token to the market in exchange for liquidity.
Hackathon theme: AI problem-solving competition
Project operation: product-driven vs attention-driven
In terms of project operation, Web2 AI projects usually adopt a product-driven route, growing and operating through product demonstrations, product function descriptions, product applicable scenarios, etc.; while Web3 AI projects often adopt an attention-driven route, with everything prioritizing the competition for market attention resources, because in the Web3 field, attention focus often represents liquidity, and attention is the most expensive asset carrier. Therefore, the previous events such as the a16z founder investing in Truth Of Terminal developers, the frequent violent remarks by ai16z founder Shaw, and the swarms developers being exposed for plagiarizing code not only did not have much impact on the AI project itself, but instead provided a boost to the development of the project and the corresponding tokens.
In the field of Web3 AI, it is difficult to establish the idea that good products bring their own traffic and speak naturally. What is believed here is whoever has the loudest voice can get more attention. Simply having a good product cannot bring a good ending to the project and the corresponding token. After all, the reality is that most Web3 AI projects are just Meme coins without any technical applications.
The so-called decentralized computing resources and decentralized data assets are nothing more than wishful thinking of project parties and retail investors.
AI is the best “graffiti pen”
Exit Mechanism: Business Model Profitability VS Token Liquidity
In terms of exit mechanism, it is the most striking, simple and crude comparison.
The exit mechanism of Web2 AI projects often depends on the profitability of the business model. Whether using AI as an automation tool, or creating an AI application, AI product, or AI large model, the ultimate goal is to attract as many users as possible, and then use subscription fees, membership mechanisms, solutions, product buyouts, advertising sales, etc. to achieve profitability and establish a stable profit business model; and most Web3 AI projects have no other exit mechanism but token liquidity. After all, the real users of these Web3 AI projects are few and far between, just like some L2 networks in the Ethereum ecosystem, which are like ghost towns.
This completely different exit mechanism also determines that the former focuses more on products, while the latter pays more attention to token assets.
Let AI manage AI, and AI serve to make money
Summary: When AI is a thing of the past, Web3 AI projects can only wait for Web2 AI technology to overflow
It is the beginning of April 2025. After experiencing two waves of AI project hype from the end of last year to the beginning of this year, the Web3 AI project briefly entered the construction period - there is no way. When market attention and liquidity are tightened simultaneously, when celebrities and presidents become the harvest sickles of cryptocurrencies, the Web3 AI project has also gone through different hype hotspots such as computing power, storage, data, AI Agent, and framework, and has become a passing thing at this stage.
In the days ahead, whether Web3 AI projects can regroup and gain more market attention and resources again may only depend on the technology spillover from Web2 AI giants, start-ups, and innovative companies. Otherwise, Web3 AI projects will still be just a concept token packaged with AI concepts. Just face the reality.
Image source:
Liangcang’s 10th anniversary, AI Hackathon competition is now underway
What determines the business ceiling is not how powerful AI is, but how close you are to AI
