Break down the five major trends under the AI ​​wave: 1 billion Agents will be on the chain in the future

Original title: 5 categories to watch for in crypto x AI agents (+ examples)

Author: 0x3van

Compiled by: Azuma, Odaily Planet Daily

AI is the main theme of this cycle, and many people firmly believe that this is a track that will continue to exist and continue to develop. Conversely, there are many legitimate criticisms that most AI agents today are perfunctory and that we will need 3-5 years before this technology becomes more meaningful.

As a preface, Crypto x AI’s project categories span multiple levels. Much of the real technological promise of Crypto x AI has to do with bootstrapping better reasoning or providing decentralized access to computing through the Crypto economy. This article on Delphi is a good start for understanding the entire stack.

However, this article focuses on our current situation with Agents. While there are exciting innovations further down the stack, Agent appears to have captured the attention of the mainstream cryptocurrency community. As Crypto x AI Agent continues to evolve, here are 5 trends to watch.

Trend 1｜Framework + launchpad: Value accumulation is beginning to become

important, and frameworks may continue to exist

Value accumulation starts to matter. Why is this happening?

Taking a step back, why would anyone hold assets? There are two main reasons why people exchange dollars for assets:

· They can sell their assets to others at a higher price as they catch up on the narrative rotation that brings in new buyers;

· The asset will generate more cash flow for them over time.

@izebel_eth once wrote in his article "Old coins are worse, new coins are better" that the only thing that matters is the flow of money. The following two reasons reflect two bullish flows:

· Growth of new holders;

· Token sinking (biased towards deflation);

However, for the most part, we don’t see anything resembling true token deflation or value accumulation. Using proxy terminals such as AIXBT via tokens is more akin to staking rather than traditional value accumulation.

This is why frameworks such as Virtuals, ai16z, zerebro, and arc have gained popularity recently. Originally primarily an investment DAO, ai16z has become one of the leading protocols of its kind since releasing details about its upcoming launchpad and token value accumulation.

Today, the framework + launchpad space is very saturated, and the first-mover effect fuels their success. There is a lot of legitimate skepticism surrounding the utility of these launchpads, as many of the proxies themselves are useless. However, those frameworks (eliza V2 + launchpad, zentients, arc and its handshake plan) have not launched their main products. If they succeed in attracting developers and users, they may continue to lead the entire field.

Why do frameworks continue to exist?

· Regardless of whether the Agent has real value, the framework used to launch the Agent will perform well because they still have "product market fit" (PMF) with speculation. Framework + launchpad allows users to have factory + casino at the same time. To a certain extent, Virtuals have replaced pump.fun in the Base ecosystem.

· More optimistically, as technology advances, leading frameworks may launch more advanced Agents, and open source software libraries such as Eliza will develop faster. Many of these launch platforms will also become the coordination layer for communication between groups and agents, which will use their tokens for some form of value transfer. For more in-depth articles on Eliza value capture, like this one Teng Yan wrote last week.

Virtuals, ai16z, Zerebro, arc are the current major players on the track, but the launchpad space is becoming increasingly saturated. The frameworks with the fastest iterations, the most scalability, and the most unique features deserve the most attention.

Trend 2 | The next round of Agents will prioritize utility and value

accumulation. DeFAI (DeFi x AI) may be the first type of Agent to achieve product-market fit.

Most agents only have meme tokens and have no utility. For the Agent industry to grow, the next round of Agents must further improve their use cases. New opportunities will come from agents that can truly accumulate value and take action. I believe that within the next year, Level 1 Agents will jump to Level 3 Agents.

We will see this first in the DeFi space. We will see more terminals that allow people to express desired results using natural language or voice, as well as agents that can perform tasks behind the scenes. Existing wallets and protocols will also integrate proxies to improve user experience.

Representative projects are as follows: Wayfinder, griffain, Hey Anon, Limitus, neur.

Trend 3｜Consumer level: entertainment agents, autonomous worlds and the

renaissance of games

Attention-driven Agents will develop more complex personalities and multimedia interactivity. This could spark the next wave of gaming and Metaverse technology.

One problem with existing agents is that they start to look more and more alike, becoming commoditized, overhyped chatbots. While some Agent projects will move towards infrastructure (many projects are already doing this), some start thinking about useful products/applications, while others may continue to go the attention token route. However, the next generation of attentional agents will be even better, developing more complex personalities and offering greater interaction possibilities. This can be achieved through audio-visual representations, or by giving the Agent a three-dimensional presence and a physical body.

There have been early attempts in this regard. Jeffy once wrote an article about implanting Zerebro into the body, and some high-profile KOL Agents were launched directly through audio-visual forms such as short videos. Slopfather and Ropirito are two early agents that adopted the video mode.

However, I think more Agent projects will add these features to make Agent more dynamic. The product fit point (PMF) in this area will be realized through the consumption layer. Individual users may want to create personalized Agent companions or interact with KOLs. This has become commonplace outside of cryptocurrencies, with some users spending hundreds of hours chatting with AI companions on sites like character.ai, a company Google recently acquired for $2.7 billion.

Additionally, these 3D Agents may find market fit within existing consumer layers, particularly games and the metaworld. Agents can add a deeper backstory to the world within a story. Imagine having an Agent act as an NPC, capable of completing tasks or playing games completely independently, with an evolving memory and personality. A truly autonomous game world may be realized through Agent.

Representative projects are as follows:

· soulgraph aims to provide tools to allow Agents to have more customized personalities and memories;

· Holoworld AI has launched Web2 partnerships with companies including L’Oréal, Bilibili and Fox to create digital avatars.

There are also some representative projects in the fields of games, metaverse, virtual world, autonomous agents and other fields:

· Hyperfy is a metaverse platform that allows anyone to create virtual worlds and uses the Eliza plug-in;

· Parallel Colony is an AI simulation game in which the Agent is the player and you can talk to them;

· Digimon is a creature game similar to Pokemon, in which creatures spiritually evolve based on interaction;

· SMOL also uses the Eliza framework. The characters in the game are agents driven by LLM, which can take actions in the virtual world according to your instructions.

Trend Four | Agentic Organizations: The Return of DAO

Decentralized Agentic Organizations are the next form of DAO evolution. Swarms or multi-Agent systems are exciting because they can coordinate and execute more advanced strategies, similar to running a company. A heterogeneous swarm composed of many different types of professional Agents/models working together may perform better than a single large model.

While fully autonomous agents and swarms may still be far away, the next iteration of DAO is likely to be about human-group interaction. This will reduce bureaucratic inefficiencies and reduce the cost and time of human execution. In the context of money flows, the next step is the revenue-earning agent, the complete revenue-earning organization.

Representative projects are as follows:

· Agent swarm infrastructure: Projects such as SwarmNode.ai and FXN are infrastructure projects specializing in multi-Agent frameworks and coordination. More mainstream Agent frameworks such as Zerebro and ai16z have also expressed their intention to build on this layer.

· DAO launchpad: Most of the early traffic we saw was focused on investment DAOs. I believe daos.fun will be the first major DAO launchpad, which spawned ai16z. There are also newer startup platforms such as daos.world, where AI-driven funds like 3BC are gaining traction.

Trend 5｜Verifiable Agents: Current Agents will develop in the direction

of being more autonomous and truly having their own liquidity.

Currently, most agents require a high degree of manual intervention. The next wave of Agents will move towards true autonomy, starting with managing their own money.

The convergence between Agent and cryptocurrency is that cryptocurrency provides a financial path for truly economical Agents. However, most agents do not control their own treasury, or leave the treasury to be managed by a human team. To achieve a truly economical Agent, the Agent must be able to manage its own funds autonomously. This allows Agent behavior to begin to evolve, because you can impose economic constraints on the Agent so that it must pay for its own reasoning costs. This introduces Darwinism to the Agent world, and they must earn income to survive.

Representative projects are as follows:

· Freysa was one of the first agents to control her own Agent and performed well (including winning Musk's attention). Recently, they just announced that they are building a framework to allow Agents to have verifiable autonomy in TEE + Agent control keys;

· Lit Protocol also has an Agent framework that allows autonomous agents to conduct on-chain transactions through the storage and execution system of private keys;

· Galadriel has launched an SDK called Proof of Sentience, which enables developers to fully verify Agents on the chain.

Conclusion: Before we see 1 billion people on the chain, we will first

see 1 billion Agents on the chain.

The user experience of cryptocurrencies is inherently unfriendly to human users. But the Agent won't care about this friction. We will start with human-Agent interaction, but the direction of cryptocurrency AI development is Agent-Agent interaction, when swarms of autonomous Agents will interact and transact on the chain and be responsible for their own economic conditions.

For agents to have economic power, enabling them to incentivize behavior (pay for services) and coordinate real activities in the real world, they need the ability to control and deploy capital. Cryptocurrency is the "home planet" of these agents - blockchain will enable agents to participate in permissionless financial activities. Stablecoins and high-performance Layer 1 are ideal tools for cost-effective, 24/7, and global transactions.

Beyond the current hype and narrative, there are good reasons to remain excited about the agent economy on the chain for the long term. Many real use cases, including DAOs and revenue-generating agents, are much closer than we think.