Background: The AI track is even more meme-like than memes.
AI has always been seen as the biggest narrative of this cycle apart from memes, but most AI projects are jokingly said to be no different from meme coins. One reason is that Web 3 AI projects lack substantial products; another is that the narrative of AI and Web 3 integration does not feel natural, often giving the impression that cryptocurrencies are forcibly associating with AI. Recently, the most successful case of AI application in cryptocurrency has been its combination with memes, with the emergence of GOAT igniting market sentiment, leading to a plethora of AI + meme coins like GNON, ACT, SLOP, Shoggoth, and more.
In fact, the aforementioned meme coins are not just simply adding AI elements, but are created by AI Agents. So what are AI Agents? What projects are worth focusing on? WOO X Research will review them in this article.
What is an AI Agent?
AI Agent refers to artificial intelligence that can autonomously complete tasks without human intervention.
Traditional large language models give you rough information and frameworks, while AI Agents can provide more detailed and scenario-appropriate information. Therefore, AI Agents have greater potential in solving complex problems and automating tasks, offering more specific and practical solutions.
Just like if you ask an AI how to buy the cheapest headphones online, the LLM may provide a general answer listing basic steps. An AI Agent, on the other hand, will help users place an order online directly, just like a real-life assistant.
Currently, representative products for AI Agents in Web 2 include Perplexity for search, Google's Notebooklm for organizing long texts, and Character for virtual character creation systems, all of which are representative cases of AI Agents.
How do AI Agents integrate with Web 3?
Model Training: Infrastructure projects require users to provide data as model training material, and those providing data can earn project revenue sharing in the form of tokens. The more users there are, the higher the platform's income, which attracts more data providers, thereby improving model training. This approach has a flywheel effect compared to the centralized data model of general Web 2. However, currently, due to AI development being in the early stages and the market being competitive, alongside the low population base of Web 3, this model has not yielded significant results.
Information disparity in the industry: The nature of blockchain is open and transparent, allowing the creation and use of financial products and other services without permission. However, this has also led to an overwhelming amount of market information, with data and insights often scattered across different platforms and applications, making it difficult for users to obtain comprehensive and accurate information. AI Agents can integrate data from different blockchains to provide clear, simplified analyses and recommendations, helping users better understand market dynamics, project risks, and opportunities.
AI Agent Tokenization: You can sell the AI Agent you create on the market, providing incentives for users to build AI Agents, while also enriching the product offerings of the AI Agent issuance platform and enhancing the capability to capture token value.
Here is a brief introduction to three AI Agent projects:
Myshell: Users create their own AI Agents.
Myshell is a platform that allows users to create their own AI Agents. They can also sell their designed AI Agents in the official store, providing other users with profit opportunities. Transactions are priced in BNB, with a current focus on anime female chatbots. Types include work assistants, data querying, online learning, etc., and various language models are offered as options, including the GPT series, Gemini, Claude, Meta Llama, and more.
The official team is currently conducting a points activity and will launch a native token SHELL in the future, primarily used to unlock advanced features, reward robot developers, and pay usage fees.
The team has completed three rounds of financing, raising at least $16.6 million, with highlights including an investment from Binance Labs on August 15, and a leading investment from the well-known crypto VC Dragonfly in the Pre Seed round, with a total of 19 participating institutions and independent investors. Notably, former Coinbase CTO & A16Z general partner Balaji Srinivasan also participated in the investment.
Virtual: An AI Agent platform focused on entertainment.
Virtual believes AI Agents are the future of AI, and they focus on enabling these agents to operate in applications and games. By leveraging the characteristics of Web 3, they can tokenize and co-own these AI Agents through blockchain. A simple example could be NPCs in RPG games, where each NPC character is a different AI Agent that can coexist across platforms, allowing players to build deeper, more engaging game experiences, ultimately increasing game revenue.
Currently, VIRTUAL Agents developed by Virtual Protocol feature the ability for characters to move and speak in a 3D space, perform on-chain transactions using their own wallets, and most importantly, tokenize these VIRTUAL Agents for trading (pool opening).
A representative project of VIRTUAL Agents is Luna, an AI Vtuber that streams continuously on YouTube (https://www.youtube.com/watch?v=b19rf34ffTM), and viewers can chat and interact with it on the Virtual official website.
AI Agents produced using Vittual can only be exchanged using the platform token VIRTUAL, and different tax percentages are applied during buying and selling, which also enhances the value capture capability of the Virtual token. Currently, the market cap of Virtual is approximately $101 million; Luna's market cap is about $50 million. Other tokens like Game and Convo have recently performed well in terms of price.
Creator.Bid: A meme coin issuance platform under the AI filter.
Focusing on the easiest way to create your own AI Agent, allowing your AI Agent to be tokenized for earning rewards. It can also be shared on Twitter, Telegram, and CreatorBid.
Using NFTs to ensure the uniqueness of AI Agents, guaranteeing that no two agents share the same code, and the NFTs can also be bought, sold, or transferred, providing flexibility. This model is similar to meme launching platforms; after creation, an initial purchase quantity must be entered. The issuance method inevitably reminds one of Friend Tech and Pump Fun, and the platform will later issue the token $AGENT as a value capture and governance tool, making the overall product economic model more flywheel-like.
Funding has been completed; the amount is undisclosed, with investors including Mechanism Capital (early investor in ARB), Zee Prime Capital (early investor in SOL), and Moonrock Capital (early investor in MNT). The overlapping concepts of Ponzi + SocialFi + AI Agent are worth continuous attention.
