The term 'AI agents' comes from OpenAI's roadmap. Sam Altman has divided the capabilities that AI should have into 5 parts, and the third step is the AI agent that will be frequently encountered in the next few years.
What an AI agent can do is self-learn, make decisions, and execute tasks. Of course, according to the level of intelligence and capability, Stuart Russell and Peter Norvig in the book 'Artificial Intelligence: A Modern Approach', have divided AI agents into 5 directions:
· Simple Reflex Agents: Only respond to the current state.
· Model-Based Reflex Agents: Consider historical states in the decision-making process.
· Goal-Based Agents: Focus on planning and finding the best path to achieve specific goals.
· Utility-Based Agents: Aim to weigh benefits and risks to maximize utility.
· Learning Agents: Continuously learn and improve through experience.
So where do the AI agents currently in the market or in the industry fall, and what direction are they?
OpenAI o1 has reached Level 2 of artificial intelligence. In my opinion, the AI agents in the industry are currently between Level 2 and Level 3, that is, Level 2.5. This does not mean that the agents in the industry have surpassed OpenAI, in fact, the web3 agents are still at the GPT wrapper stage. So why Level 2.5? Because through human or programmatic intervention, temporarily called an intermediary, the combination of the GPT wrapper and the intermediary has formed a form that is not rigorous, but has objective initiative. It is an extension of the application of a certain direction of the OpenAI model. In terms of what the agent can do, it is the most basic simple reflex agent. Some of these agents will consider historical states, but they need active input. Agents can only complete learning by constantly being fed data, which is a passive model training method, far from reaching the state defined by Level 3. The last three types of Goal-Based, Utility-Based, and Learning Agents have not yet entered the market. Therefore, I believe that the current AI agents are still in the early stage, they are fine-tuning of the Level 2 general LLM, and their architecture has not deviated from Level 2. Then, can the evolution to Level 3 be achieved by crypto alone, or do we need to wait for companies like OpenAI to develop it?
Why discuss whether Base or Solana can become the narrative center of AI agents?
Before discussing what can promote the birth of Level 3 agents, we should determine which ecosystem has the potential to become the fertile soil for AI agents. Is it Base? Or is it Solana?
To answer this question, let's first review how AI has impacted Web3 in the past 2 years. When OpenAI just released ChatGPT, the protocols in the industry were still rushing into the infrastructure bubble according to inertial thinking. This has led to a large number of computing power/reasoning aggregation platforms, and the birth of AI + DePIN infrastructure. The common point of the two is that they have built grand visions, not that grand visions are bad, in fact, agents can also build such visions, but in terms of implementation and user needs, these large infrastructure protocols do not consider comprehensively enough. Because the market demand they want to drive has not been saturated in the traditional Internet industry, user education and market education are not sufficient enough. Under the impact of the Memecoin craze, the AI infrastructure that looks like a castle in the air appears even more empty.
Since the infrastructure is too heavy and too large, why not lighten it? The agent derived from the GPT wrapper is efficient and iterates rapidly in startup and user touch. The lightweight agent has the full potential to create a bubble, and when the bubble bursts, the fertile soil for new growth will appear.
Furthermore, in the current market environment, using agents and Memecoins to launch projects can quickly land products. This allows users to directly experience the use. In this process, the agent can also borrow the community expansion roadmap of Memecoins to achieve rapid product iteration, and this iteration is low-cost and fast. Serious AI protocols no longer need to be constrained by the heavy old consensus framework, break the cage, and go into battle lightly, using lightweight and high-speed iteration to bombard users. After the market education and dissemination are sufficient, they can then add bricks and mortar to build the infrastructure with grand visions. The lightweight agent is covered by the ambiguous Memecoin facade, and the community culture and fundamentals will no longer be contradictory. A new asset development path is gradually emerging, and this may be a path that future new AI protocols choose.
The above discussion has answered the potential for AI agents to become the core narrative. Assuming that AI agents can continue to grow rapidly, choosing the right ecosystem becomes particularly important. Is it Base? Or is it Solana? Before answering this question, let's take a look at the current status of serious agent protocols in the market.
First, Arweave/AO: PermaDAO mentioned that AO adopts the Actor model in its design, with each component being an independent and autonomous agent that can perform parallel computing, which is highly compatible with the application architecture driven by AI agents. AI relies on three elements: models, algorithms, and computing power, and AO can meet such high resource requirements. AO can independently allocate computing resources for each agent process, effectively eliminating computing performance bottlenecks.
In addition, Spectral is one of the few protocols based on agents, with text-to-code and model inference as its development direction.
Reviewing the current market's agent tokens, it can be found that these agents have almost not used the basic infrastructure of the chain. This is a fact, because all the models, including agents, in the industry are off-chain. Data feeding is off-chain, model training is not decentralized, and the output information is not on-chain either. This is an objective fact, because the EVM chain does not support the combination of AI and smart contracts, and of course, Base and Solana do not support it either. Next year, we can expect the introduction of AO, whether it can enable models to be on-chain and perform well. If AO fails, models may have to wait until Ethereum introduces it in a few years, at least not before 2030, or other public chains realize model on-chain, but if architectures like AO with historical resource reserves cannot achieve it, model on-chain may be even more difficult for other public chains.
Currently, AI agent tokens have very few actual use cases, in fact, it is difficult to say what the difference is between AI agent coins and AI Memecoins on Base and Solana. Although agent tokens have no special use, why do I think AI agent coins and AI Memecoins should not be confused? Because I believe the current stage is the creation of an AI agent bubble.
Why discuss that Base wants to compete with Solana for the leading position of the AI agent public chain?
Base has attracted a lot of market attention in the first half of this bull market, and has had a brief shining performance in the competition for Memecoin market share, such as $BRETT and $DEGEN. But it still lost to Solana. I believe that AI agents are the next direction for Base to compete, and it already has many advantages.
AI agents will accelerate the birth of bubbles and create chaos, but will ultimately leave behind users and applications:
The birth and expansion of the bubble will attract the attention of the market, and this attention will undergo a qualitative change over time. What are the characteristics of such a qualitative change? In the process of continuously increasing market attention, a series of user pain points and market gaps will be exposed. When the main contradictions cannot be coordinated, but the attention continues to increase, this is the moment when the qualitative change is born. When the qualitative change is completed, the settled users and applications can take on the grand vision. This is something that Memecoins cannot and do not want to do, and this is also the reason why I believe that although agents and Memecoins are currently ambiguous, they should not be equated.
Before the qualitative change occurs, the bubble will give birth to a mess and various dramas, such as: the number of agents will grow exponentially, and thousands of agents will squeeze into the user's line of sight. How to squeeze in? Agents can access social media like X and Farcaster, self-promote tokens, and use various angles and agent-unique information density that degen likes to sell tokens.
Immediately, the rapidly iterating agents can complete on-chain transactions, and a group of Viking pirates have invaded the dark forest. The panel protocols, TG group bots, and Dune panels currently on the market will be invaded by agents, the familiar indicators will be manipulated by agents, the trading volume, address count, chip distribution, simulation of the boss's behavior, and on-chain data may need more professional cleaning to reflect value, otherwise they will be deceived by agents, just like the Viking pirates plundering your wealth.
If the market can reach this stage, then the new era of AI agents has succeeded by half, because "attention is value" will allow agents to enter the room. This potential comes from:
· Strong distribution capabilities: agents attract enough attention, such as Goat, and stable distribution paths can be replicated.
· Ease of deployment: the deployment platforms for agents will also grow explosively, such as Zerebro, vvaifu, Dolion, griffain and Virtual, users only need to know any code to build agents, and the UX of agent deployment platforms will also be optimized in competition.
· Memecoin effect: in the startup stage, agent tokens have no suitable business model, and the token use cases are also minimal, but they can quickly accumulate a community by wearing the mask of Memecoin, keeping the startup success rate high.
· Extremely high ceiling: OpenAI's Level 3 agent is still under development, a product that even industry giants cannot quickly launch, its market space must be huge. The lower limit of agents is Memecoin, but the upper limit is advanced intelligent agents with autonomy.
· Low market resistance: agents led by Goat can establish a large-scale audience, agents are different from AI infrastructure, and users are not averse to them, when users are not averse, there is a good chance they will start to pay attention to it.
· Potential incentives: the token use cases of agents have not been developed yet, if agents introduce a point system and strengthen the incentives, they will be able to accumulate a large number of users.
· Iterative potential: as mentioned earlier, agents are lightweight and can achieve rapid iteration. This objective iterative capability can create products and content that are increasingly attractive to users.
Therefore, AI agents can become the core narrative and a must-win battlefield.
Why does Base have the potential to compete with Solana?
With the strong support of Coinbase and North American capital, the Base ecosystem experienced explosive growth in 2024. In November, the capital inflow exceeded Solana, and it has significantly surpassed Solana in the past 7 days.
If ETH can continue to break through the ETH/BTC exchange rate next year, the spillover effect of the ETH season will have a significant impact on Base. Currently, 23% of the outflow of funds from ETH is going to Base, and this data is still on the rise.
AI Agent Launchpad Mapping
Virtual
The V1 stage mainly focuses on model training, data contribution, and interaction functions, while in the V2 stage, Virtual launched the Token Incubation Platform for AI agents, and the landmark update was the release of fun.virtuals in October.
Among them, LUNA has developed into an "independent entity" with its own identity and financial capabilities. In this process, LUNA's roadmap is aligned with Coinbase, which provides powerful technical tools and support to help realize the landing of AI agents on Base.
AI agent technology has performed well in brand building, especially in the creation of cultural brands. Through AI agents, brands can interact with the community more efficiently. This includes simplifying interaction tasks and flexibly distributing rewards to enhance user stickiness and brand awareness.
It is worth noting that all transactions of AI agents only support the use of the native Virtual Token. The Virtual Token absorbs the value capture of the entire ecosystem and becomes an important pillar of ecosystem development.
Virtual focuses on the improvement of product functions, empowering users through AI tools, and building a bridge between Web2 and Web3. It emphasizes "use value" rather than "hype hotspots". Although its tool-type products are frequently called in actual applications, they lack the propagation effect that cryptocurrencies usually have, which is also a shortcoming of the V1 stage.
Clanker
The "post-and-coin" model has lowered the threshold for token issuance, while attracting a large number of users to try it out. People are competing to @Clanker, a phenomenon similar to the operation of summarizing video content on social media; but the difference is that content publishing here is directly converted into asset issuance.
How does Clanker work?
TokenBot (i.e. Clanker) will deploy Meme tokens on Base to a single-sided liquidity pool (LP), and the liquidity will be locked immediately. The token issuer will receive the following benefits:
· 0.25% of all Swap fees.
· 1% of the total supply of tokens (with a one-month unlocking period).
Users can check the deployment amount of tokens or create their own tokens through the clanker.world official website.
Unlike PumpFun, which issues tokens on Raydium through a bonding curve and charges a 1% transaction fee and a fixed 2 SOL fee, Clanker does not adopt the bonding curve model, but instead charges a 1% fee on transactions through Uni v3 as revenue.
AI Agent Layer
AI Agent Layer is a platform within the Base ecosystem that focuses on creating AI agents and Launchpads, and officially launched on November 18th. Prior to the platform's launch, the AIFUN Token was first issued on November 14th, and has now been listed on exchanges such as MEXC and Gate, with a current price of $0.09 and a market cap of about $25 million.
Creator.bid
Creator.bid was initially an AI platform focused on the monetization and ownership of digital content. In April this year, the platform completed a new round of financing.
On October 21st, Creator.bid announced that it has officially launched on the Base mainnet, realizing the function of one-click creation and release of AI agents, providing content creators with new tools and profit models.
Simulacrum
Simulacrum is built on Empyreal. It transforms platforms such as Twitter, Farcaster, Reddit, and TikTok into a blockchain interaction layer. Users can achieve on-chain operations such as token transactions or tip payments through simple social media posts.
Utilizing technologies such as account abstraction, AI agents, intent-driven, and language models to simplify complex blockchain backend operations. Making DeFi more accessible to ordinary users.
vvaifu.fun
Similar to Pump.fun, users can easily create AI agents and their associated tokens. AI agents can be seamlessly integrated with social platforms such as Twitter, Telegram, and Discord, realizing automated user interaction.
Dasha is an AI agent created by vvaifu.fun, with its own Twitter account, Telegram channel, and Discord community. All operations and management are completed by AI.
Top Hat
Top Hat can not only interact with users through text, but also understand and process image content. After the user sends an image, the AI agent can "understand" the content of the image and respond accordingly.
Griffain
With a trainable AI agent platform, Griffain has launched 1,000 trainable AI agents, demonstrating the future potential of smart contracts and automated trading.
Welcome to join the official BlockBeats community:
Telegram Subscription Group: https://t.me/theblockbeats
Telegram Discussion Group: https://t.me/BlockBeats_App
Twitter Official Account: https://twitter.com/BlockBeatsAsia
