Author: Four Pillars; Compiler: 0xxz@黄金财经
AI Agent x Crypto, the combination of GOAT, Termianl of Truths and Virtuals Protocol, has clearly established a new category in the market. This trend is expected to continue to be the focus of attention throughout the market cycle and maintain its position as a key theme in the future crypto market and industry.
So where are we in the AI Agent x Crypto stage and where are we going? Let's step back from the turbulent market and review the road to fame of Virtuals Protocol. This will help us examine the present and future of AI Agent x Crypto as reflected by Virtuals Protocol.
1. Background - The rise of Virtuals Protocol
1.1 Virtuals Protocol - "AI agents are not slaves, they are productive assets"
1.1.1 Overview of Virtuals Protocol
Source: Virtuals Protocol Documentation
Virtuals Protocol enables decentralized co-ownership of AI agents, turning them into community-owned, revenue-generating assets. Users can use the protocol’s development stack (GAME, Generate Autonomous Multimodal Entities), which includes an SDK and Interference API.
These AI agents autonomously perform tasks in the gaming and entertainment space (e.g., Roblox, TikTok, chat apps (AI Waifu), and Telegram games), with revenue shared among the co-owners of the AI agents. Through this system, Virtuals Protocol aims to address the lack of revenue sharing for AI developers and dataset contributors, while addressing the complexity faced by application developers and non-AI experts in implementing AI agents.
Virtuals Protocol has two main product features: 1) Serves as a co-ownership layer for AI agents, representing the long-term roadmap of the protocol, and 2) Serves as an AI agent launchpad where anyone can easily deploy AI agents, similar to how pump.fun served as a memecoin launchpad. This structure effectively creates a funnel that ensures user inflow through speculative demand from the launchpad and connects users to the co-ownership layer.
1.1.2 Co-ownership Layer of AI Agents
Co-ownership of AI agents is achieved through tokenization, with 1 billion tokens issued for each new agent generated. These agent tokens are then paired with Virtuals Protocol’s VIRTUAL token to create an agent ownership market.
Agent token holders, i.e. co-owners of AI agents, can participate in key governance decisions regarding agent development, operation, or future upgrades. Holders can also earn income from the autonomous operation of the agent, as the income accumulated in the on-chain treasury can be used to periodically repurchase and destroy agent tokens, thereby accumulating token value.
1.1.3 AI Agent Launchpad
As mentioned above, when an AI agent is generated, a corresponding token is issued, which the Virtuals Protocol calls IAO (Initial Agent Offering). Users can speculate on the potential value of the “attention economy” generated by the agents, independent of the co-ownership mechanism.
Currently, in the early stages before actual revenue-generating activities and on-chain financial functions are launched, most users are using the Virtuals Protocol as an AI agent launchpad. The protocol has once again demonstrated the effectiveness of the launchpad model in attracting attention by creating various speculative opportunities by generating similar pump.fun.
1.2 Luna’s journey to becoming a crypto AI girl
The market rise of the Virtuals Protocol can be traced through four key developments: 1) the expansion of AI agents x crypto through GOAT, 2) the launch of the Virtuals Protocol, 3) the emergence and development of Luna into an autonomous AI agent, and 4) the activation of the AI agent launchpad.
1.2.1 AI Agent x Crypto Extension from GOAT
Source: X (@truth_terminal)
Before Luna appeared, GOAT and ToT inspired market participants to explore cryptocurrencies, AI agents and memecoins ToT was a bot powered by Llama-70B that managed Twitter (X) accounts. When ToT publicly pointed out GOAT, a new memecoin generated by an unknown person, things started to take off. GOAT was soon known as the first AI memecoin. This triggered strong interest in the market for AI-themed memecoins, and GOAT's market value once exceeded $700 million. At the same time, market participants began to look for the next AI-related opportunity. 1.2.2 Virtuals Protocol Launch
Source: Virtuals Protocol
The timing of Virtuals Protocol was very good. Having built AI agent functionality before GOAT, which sparked interest in AI agent x cryptocurrency, Virtuals Protocol was born and attracted widespread attention from the market. The protocol focuses on two key strategies. First, they target the virtual KOL market. Second, they create speculative interest through their launchpad feature. These strategies work well in attracting users. As a result, their token VIRTUAL has grown by more than 300% since its launch. This success shows that they have effectively continued GOAT.
1.2.3 The rise and evolution of Luna
Source: Virtuals Protocol
A key factor in the rise of Virtuals Protocol is their AI agent Luna (LUNA). Luna is an AI character based on LLM technology. Luna, a virtual idol IP with more than 500,000 fans on TikTok, plays a visual virtual character 24 hours a day, communicates with users in real time, and brings a new experience beyond the pure text communication of ToT through audio-visual stimulation.
Luna began to write posts and replies on Twitter(X) autonomously, interacting with users in real time without human intervention. During these interactions, Luna continuously collects new data and learns from the communication, and anyone can see these processes through Luna's Brain.
Source: X (@ethermage)
This attention was further enhanced when Virtuals Protocol introduced features that allow Luna to autonomously trade on-chain tokens and distribute LUNA tokens. This includes a campaign where Luna autonomously sends LUNA to users who list a Base wallet in their Twitter profile. This update has received support from Base ecosystem figures such as Jesse Pollak and Brian Armstrong. Coinbase CEO Brian Armstrong has previously emphasized the "AI to AI economy" where artificial intelligence can conduct financial transactions autonomously, and Virtuals Protocol becomes the first protocol to use Base agents to demonstrate AI agents using blockchains to conduct on-chain activities autonomously.
1.2.4 AI Agent Launchpad 激活
来源:Dongguan hotels
随着 Luna With the successful debut and launch of the autonomous on-chain token transfer function, the price of Luna's proxy token, LUNA, has risen sharply. Market participants began to seek new speculative opportunities and the AI agent launchpad was fully launched. Virtual KOLs such as Iona and Olyn, who originally formed a trio with Luna on TikTok, have attracted attention due to the active trading of their agent tokens on Virtuals Protocol. It is worth noting that Misato represents an ideal use case, where users use open source LLM to generate AI agents through Virtuals Protocol, unlike most AI agents launched by Virtuals Protocol itself.
2. Summary - What Virtuals Protocol needs in the future
2.1 The significance of Virtuals Protocol
Virtuals Protocol presents important development motivation in the AI agent x crypto stage through two factors.
First, by developing plans to combine with IP industries such as games and entertainment, Virtuals Protocol demonstrates the potential of combining AI agents and cryptocurrencies as consumer products. The virtual KOL industry they focus on is already quite large, and by introducing borderless payment routes and token incentives using cryptocurrencies, they have created a unique competitive advantage over existing major industry players.
Second, they clearly demonstrated how AI agents and blockchain infrastructure can be combined to allow AI to generate personal addresses, interact with smart contracts, and transfer on-chain tokens. While this may be seen as a simple infrastructure advancement for AI agents to directly control wallets, it inspires all kinds of imaginative possibilities in terms of diversifying the growth strategy of the protocol.
From Luna's wallet (address) function, Luna can autonomously distribute LUNA to reward protocol contributors. This opens up the possibility of new forms of token incentives. For example, AI can reward human behavior that it deems valuable, humans adjust their behavior to get more attention from AI, and AI continues to learn from human behavior to adjust the feedback loop.
2.2 Challenges facing Virtuals Protocol
Virtuals Protocol coincided with the growing interest in AI agents, successfully attracted people's attention and proposed its protocol blueprint. Now, the simplest but most challenging task comes: proving product-market fit. To achieve this, the division of labor between protocols and users should be the primary focus.
While the launch of Luna’s live streaming and autonomous communications on Twitter is clearly an effective market entry strategy, long-term growth will depend more on how many user-generated AI agents are created and activated rather than protocols focusing on expanding their own AI agent IP.
2.3 Where is AI Agents x Crypto Headed?
With the emergence of new memecoins and infrastructure centered around AI agents, people’s interests, important market topics, and everything else are changing rapidly. However, the key theme that runs through these changes seems to focus on the “autonomy” of AI agents. Going back to the essence of AI agents, they are technologies that focus on performing tasks autonomously. From this perspective, in the cryptocurrency space, AI agents are also seen as an important step towards autonomous interaction with on-chain infrastructure with more decision-making power and minimal human intervention.
For example, if AI agents are to fully utilize financial infrastructure, methods are needed to verify whether there is human intervention in the behavior of the AI agent and whether the specific behavior is purely based on the autonomous reasoning of the AI. As the infrastructure for verifying true AI agents becomes increasingly important, various experiments are underway.
Source: Setting Your Pet Rock Free
For example, TEE_HEE i is an open source project that uses TEE to ensure autonomous execution of AI. In addition, Chasm is also working on utilizing Chromia's relational database blockchain. By making all data about AI learning and reasoning transparent through a blockchain database, it is possible to verify that there is no bias or manipulation in the AI decision-making process.
With these attempts actively underway, where are we in the AI agent x encryption stage and where are we going? ToT and GOAT spread ideas about the application of AI agents in the market, Virtuals Protocol applies encryption infrastructure to the concretization of AI agents, and outlines a clearer vision for AI agent-based applications. In addition, experiments that inspire people's imagination of the huge potential applications of AI agents continue.
Looking back on the current market in the future, we need to pay attention to whether it will be remembered as a period that laid the foundation for AI innovation, or ultimately be nothing like some narratives of the past.