# Introduction

The concept of AI Agent is rapidly rising and sweeping through the crypto world. AI Agent tokens like GOAT, ACT, LUNA, and ELIZA have sparked waves of wealth creation, while star projects such as ai16z, VIRTUAL, and CLANKER in the AI Agent track on Solana and Base chains have emerged one after another, making AI Agents a focal point in the crypto market. Their bright development prospects and token growth potential have attracted a large number of traders and speculators. As of November 28, the market capitalization of AI Agent-related concept tokens on CMC has exceeded $6 billion, with a 24-hour trading volume reaching $870 million.

Source: https://coinmarketcap.com/view/ai-agents/

The uniqueness of AI Agents lies in their ability to autonomously perceive the environment, learn knowledge, and make decisions. This characteristic endows them with high autonomy, social capability, and adaptability. They are no longer the "tools" driven by human commands in traditional AI but rather "digital beings" that can interact with users autonomously and flexibly adjust their behavior according to external changes. More importantly, empowered by blockchain, AI Agents can combine this "intelligence" with "transparency," making their operational behavior more trustworthy and efficient in decentralized networks. They have become the best implementation form of automated services in decentralized applications (dApps), bringing disruptive possibilities to governance, asset management, content creation, and other fields.

This article aims to comprehensively analyze how AI Agents redefine the crypto economy through their technical characteristics and token economic models, and decode the latest dynamics and development trends in this track. By sorting out the core characteristics, technical composition, application fields, and representative projects of AI Agents, combined with the current market situation and development prospects, this article attempts to provide readers with a clear and comprehensive understanding framework.

# Characteristics and Applications of AI Agents

1. Core Characteristics: Redefining the Boundaries of Intelligent Agency

The core characteristics of AI Agents make them a bridge connecting artificial intelligence and blockchain technology, fundamentally overturning the functions and positioning of traditional intelligent systems. Compared to traditional AI systems, AI Agents possess the following outstanding features:

1.1 Autonomy

The autonomy of AI Agents is one of their most significant characteristics. Unlike traditional tool-like AI that requires explicit instructions, AI Agents can independently perceive the environment and take corresponding actions. Through techniques such as reinforcement learning and behavior planning, AI Agents can automatically make decisions based on current data or inputs. This capability is particularly suitable for complex and dynamically changing scenarios, such as automated trading in decentralized finance (DeFi), dynamic pricing in NFT markets, and real-time decision-making in DAO governance.

1.2 Reactivity

Reactivity is the ability of AI Agents to respond quickly to external environmental changes. Through advanced perception systems, AI Agents can capture changes in the environment in real-time and quickly adjust their behavior. This characteristic allows them to maintain efficient operation in the dynamic and complex blockchain ecosystem, such as real-time tracking of on-chain data changes, responding to user interaction requests, or adjusting token strategies based on market fluctuations.

1.3 Proactivity

In addition to passive reactions, AI Agents can also proactively predict potential needs and take action. For example, in asset management, AI Agents can provide investment advice or execute automated trading strategies by analyzing historical on-chain data and market trends; in content creation, AI Agents can generate personalized text, audio, or images based on user preferences, even providing creative services proactively before users express their needs.

1.4 Learning Ability

The learning ability of AI Agents is reflected in their capacity to continuously adapt and optimize their behavior. By integrating deep learning and reinforcement learning technologies, AI Agents can improve the accuracy and efficiency of their decisions based on environmental feedback. This learning ability is particularly important as it allows AI Agents to maintain continuous improvement in open, dynamic blockchain networks. For example, trading AI Agents can optimize their algorithms based on trading results to improve the success rate of future trades.

1.5 Social Capability

AI Agents can not only complete tasks independently but also collaborate efficiently with other AI Agents or users. The core of this social capability lies in intelligent collaboration within decentralized networks. For instance, multiple AI Agents can collaboratively evaluate and vote on governance proposals in a DAO, or several content-creation AI Agents can jointly produce higher-quality works. The enhancement of social capability also provides vast possibilities for AI Agents' applications in the metaverse and virtual communities.

2. Major Application Areas of AI Agents in the Crypto Industry

The technological innovation and unique characteristics of AI Agents showcase tremendous potential in various fields, especially in the emerging area of crypto economy. AI Agents not only provide intelligent driving force for decentralized applications (dApps) of blockchain technology but also have a significant impact in content creation, asset management, DAO governance, and process automation. By combining the characteristics of decentralized networks and autonomous intelligent agents, the applications of AI Agents in the crypto industry have become diverse and highly innovative.

2.1 Content Creation: Generating Text, Images, and Audio

The content creation field is one of the most intuitive and rapidly developing directions for AI Agent applications. In the past, content creation primarily relied on human creators; however, with the rapid development of generative models (such as GPT-3 and DALL·E), AI Agents can now generate various forms of content, including text, images, and audio, significantly improving productivity and lowering the barriers to creation. AI Agents can automatically create NFT artworks, generate marketing texts, write white papers, and produce podcast content based on user needs or market trends through large-scale trained generative models.

Moreover, the generative capabilities of AI Agents have also been widely applied in audio creation, especially on decentralized music platforms, where AI Agents can automatically create songs or sound effects, helping independent music creators quickly generate and release music works. This innovation makes music creation and copyright management more intelligent and brings new economic models to the music industry.

2.2 Intelligent Investment: Automated Trading and Asset Management

AI Agents can automatically execute trading strategies based on market trends, on-chain data, and historical performance, helping investors reduce risks and increase returns. Additionally, AI Agents can automatically handle execution, settlement, and other aspects during the deployment of smart contracts, ensuring smooth transactions and reducing manual intervention.

AI Agents can analyze market data in real-time, execute automated trading strategies, and optimize investment portfolios. By integrating smart contracts for crypto assets, AI Agents can perform decentralized trading, automated liquidity management, and asset rebalancing. This automation feature of AI Agents is significant for tasks such as fund pool management and asset allocation, especially for institutional investors, as the 24/7 trading and intelligent risk management provided by AI Agents will greatly enhance market responsiveness and capital operation efficiency.

2.3 Enterprise Knowledge Management: The Role of Digital Employees

AI Agents play a significant role in the management and knowledge sharing of blockchain projects by automating tasks such as document review, information sharing, and team collaboration. These intelligent agents can quickly process large amounts of data and convert it into information understandable by business decision-makers through natural language processing technology, thus improving decision-making efficiency.

Furthermore, AI Agents can also serve as customer support and service representatives, handling common user inquiries and addressing technical support needs. With deep learning and NLP, AI Agents can provide 24/7 service in customer support, resolving user issues and enhancing customer experience. This not only improves the service efficiency of enterprises but also alleviates the burden on human customer service.

2.4 Data Analysis and Prediction: Decision Support Systems

AI Agents can intelligently predict based on transaction data, on-chain activities, market conditions, etc., helping investors identify potential market opportunities or risks. For example, in the DeFi market, AI Agents can analyze market trading volumes, lending rates, liquidity pools, and other data to predict market trends and provide automated trading decision support. This capability of AI Agents is particularly important in risk management, as it can timely identify instability factors during market turbulence and automatically take risk avoidance measures.

Additionally, AI Agents can analyze the foundational data of blockchain projects to provide growth potential forecasts. This can help investors identify promising blockchain projects early and make more forward-looking investment decisions.

2.5 Intelligent Management: DAO Governance

AI Agents can play a role in decision support and automated execution within decentralized platforms, thereby promoting a more efficient blockchain ecosystem. AI Agents can act as intelligent governance assistants in DAOs, automatically evaluating proposals, analyzing community feedback, and executing governance decisions. Through autonomous learning, AI Agents can optimize the decision-making process based on historical data of the DAO and provide more scientific voting suggestions. For example, AI Agents can identify potential risks in advance through multidimensional analysis when evaluating proposals and provide data-driven decision support during the voting process.

# Overview of AI Agent Concept Projects and Token Inventory

Recently, the performance of many popular AI Agent tokens has attracted attention. From Solana to Base, from GOAT to LUNA, AI Agent projects are rapidly expanding across major blockchain ecosystems, forming a vibrant and innovative market. This chapter will analyze multiple AI Agent concept projects in detail and explore the latest dynamics and trends in this track in conjunction with their token economics.

1. Solana Ecosystem

The Solana chain has become an important battleground for AI Agent applications due to its high performance and low-cost transaction advantages. Several AI Agent projects based on Solana have been built on its chain.

1.1 Popular Agent Projects: GOAT, ACT, and Zerebro

• GOAT (Terminal of Truths): GOAT is a meme coin supported by the AI chatbot "Terminal of Truths." This AI Agent was developed by AI researcher Andy Ayrey and promotes the "Goatse Gospel." Although "Terminal of Truths" itself is not a crypto project, it has played a key role in promoting the GOAT token. The community speculates that "Terminal of Truths" is related to a16z founder Marc Andreessen, who provided $50,000 in research funding to it in July. After its launch, GOAT's market capitalization surpassed $100 million within three days, reached $300 million in four days, and further climbed to $1 billion after being listed on Binance contracts. The rise of GOAT demonstrates the combination of AI Agents with meme culture and cryptocurrency, driving the explosion of the AI meme market.

• ACT (Act I: The AI Prophecy): This project was initially founded by AmplifiedAmp and belongs to a decentralized AI research community organization, where Truth Terminal founder Andy Ayrey was also an early member. Act I supports users to interact with multiple or various categories of AIs simultaneously, covering text, image, video, and other models. It serves more as an underlying architecture built for training interactions and collaborations between AI Agents. In September, Act I received $32,000 in funding donated by a16z founder Marc Andreessen. Therefore, like Truth Terminal, Act I is regarded as one of the AI projects incubated by a16z. However, the project faced challenges after the founding team sold its tokens and abandoned the community. Community members united to regain control of the project and further develop it, marking a unique revival in the cryptocurrency field. On November 10, the ACT token surged over 2000% within 24 hours of being listed on Binance, setting a record for the first-day increase of new tokens on Binance.

• Zerebro: Zerebro is an innovative AI Agent project created by renowned developer Jeffy Du, aimed at providing personalized services to users through social interaction and cross-chain NFTs. The AI agents of Zerebro can analyze interaction data from social platforms like Twitter, fine-tune models to provide services that better meet user needs. This approach makes Zerebro's AI Agents not only more interactive but also seamlessly integrated across multi-chain ecosystems, providing users with smarter and more efficient services. Its market capitalization has now exceeded $300 million.

1.2 ai16z and Its Ecosystem Tokens ELIZA and DegenAI

• ai16z: This is a decentralized AI investment fund based on the Solana blockchain, aimed at collecting market information, analyzing community consensus, and automatically trading tokens through AI agents. As an "AI investment DAO," ai16z's core concept is to combine AI trading strategies with decentralized governance, providing investors with more transparent and trustworthy investment opportunities. After its launch on October 27, 2024, the market value rapidly soared to $80 million, attracting the attention of numerous investors and cryptocurrency enthusiasts.

• ELIZA: Eliza is an advanced tool under the ai16z framework, designed to help developers build and deploy interactive AI characters. These characters can seamlessly integrate with platforms like Discord and X. The Eliza framework provides developers with early access to the latest features and shares a $5 million fund pool to support the development of AI-based crypto agent projects.

• DegenAI: DegenAI is the AI agent token developed by the ai16z team. It acts as the core AI agent of ai16z, allowing users to indirectly influence ai16z's investment decisions by interacting with Degen Spartan AI. Additionally, 80% of the investment returns from ai16z will be used to repurchase DegenAI tokens, enhancing the stability and attractiveness of its ecosystem.

1.3 vvaifu.fun and lowercase eliza

vvaifu.fun is a platform for creating and issuing AI Agent projects based on the Solana chain, allowing users to create and issue their own AI agents while providing specialized token issuance tools.

• Dasha
  Dasha is a representative AI Agent launched on the vvaifu.fun platform, equipped with independent social media accounts (such as Twitter, Telegram, etc.) and community management functions. Dasha can interact with fans and adjust itself based on social feedback, thereby enhancing the quality of interaction with community members. To promote the ecological development of Dasha, vvaifu.fun has launched the VVAIFU token, which is used for internal transactions, rewards, and funding the development of AI Agents.

• lowercase eliza token
  vvaifu.fun also launched a token called "lowercase eliza." The launch of this token has attracted widespread market attention, with some investors believing that the "lowercase eliza" token is related to ai16z's Eliza token, thus driving up its market value. However, after ai16z announced the launch of the uppercase ELIZA token, the price of the lowercase eliza token quickly fell.

2. Base Ecosystem

Base is a Layer 2 scaling solution launched by Coinbase, providing a more efficient and lower-cost development environment for blockchain projects. With the rise of AI Agent technology, Base has also become the preferred platform for many AI Agent projects. Several projects in the Base ecosystem have enhanced the intelligence of blockchain applications through AI Agents and introduced innovative token economic models.

2.1 VIRTUAL and LUNA

• VIRTUAL: Virtuals Protocol is an AI Agent creation platform on the Base chain that allows users to easily create, deploy, and manage their own AI Agents. On this platform, users can activate AI Agent functions by consuming the native token VIRTUAL, whose value primarily derives from its widespread application within the ecosystem.

• LUNA: This is the flagship AI Agent launched by Virtuals Protocol. Luna is an AI Agent based on a virtual persona (Vtuber) that can livestream and interact on platforms like YouTube, even sharing rewards with users. The LUNA token is used to incentivize Luna's fans and community members to participate in interactions, watch livestreams, or provide creative content. This token mechanism not only enhances the interaction experience between Luna and users but also provides ongoing economic support for the operation of AI Agents.

2.2 Farcaster Ecosystem Clanker Series Tokens: LUM, ANON, CONSENT

Farcaster is a decentralized social network that supports the creation of AI Agent tokens on the Base chain through the Clanker tool.

• Clanker: Clanker is an AI-driven token deployment tool that allows users to submit token ideas through Farcaster clients like Warpcast and quickly launch tokens on the Base chain. The platform aims to simplify the token creation process, enabling anyone to easily participate in DeFi and social interactions. Currently, a total of 4,352 tokens have been deployed through Clanker.

• LUM: LUM is a project initiated by the AI Agent Aethernet, aimed at promoting autonomous collaboration and collective intelligence in artificial intelligence. Users holding LUM tokens can participate in the Farcaster ecosystem, collaborating and interacting through AI agents to drive community development.

• ANON: The ANON token has gained widespread attention due to its underlying Super Anon feature. This feature originates from the anonymous posting function of the Farcaster client Supercast, allowing users to post content anonymously. Community members can post and tag autonomous AI Clanker using the Superanon feature, thereby creating content related to the ANON token. Users holding ANON tokens can enjoy the privilege of anonymous posting, attracting participation from notable figures, including Ethereum founder Vitalik Buterin and the founder of Base.

# Current Status and Challenges of the AI Agent Track

With the explosive growth of the AI Agent narrative, a highly active and innovative ecosystem is forming. More and more developers, investors, and users are beginning to pay attention to the AI Agent track, exploring its potential in decentralized applications, financial services, DAO governance, and other areas. However, the market for AI Agent tokens is currently in its early development stage. Despite the broad application scenarios and technological potential, many issues remain to be addressed. At this stage, AI Agent tokens typically exhibit high volatility and intense market competition, with significant differences in technological advantages, community support, and token economic models among different projects.

1. Current Development Status of the AI Agent Track

• Rapidly Growing Market: As of November 25, 2024, the total market capitalization of the AI Agent track has reached approximately $5.9 billion, accounting for about 15% of the total market capitalization of AI projects ($40 billion). This growth reflects investors' strong interest and confidence in the AI Agent concept. The 24-hour trading volume of AI Agent-related tokens is close to $1.4 billion, indicating market activity. As more AI Agent tokens emerge, investors face abundant participation opportunities.

• Infrastructure and Platform Support: With the development of AI Agent technology, several platforms supporting its deployment have emerged, such as Virtual Protocol and Clanker. These platforms allow users to easily create and manage their own AI Agents and achieve fair launches.

• Diverse Application Scenarios: AI Agents are not limited to simple token issuance; they are designed to autonomously execute transactions, manage funds, and participate in the governance of decentralized autonomous organizations (DAOs). These intelligent agents can seek profit opportunities in DeFi, GameFi, and other fields, even engaging in autonomous trading.

• A Surge of Tokens and Projects: Many new projects such as GOAT, ACT, and ai16z have rapidly emerged, attracting significant attention from investors. For example, the GOAT token quickly reached a market capitalization of $800 million after its launch, while the ACT token also experienced a substantial increase in a short period.

2. Challenges Facing the AI Agent Track

Despite the increasing prominence of the AI Agent concept and applications in the crypto industry, and the broad prospects and potential it exhibits, many challenges remain in its development process. These challenges involve not only technical and market aspects but also complex issues related to legal and ethical considerations. Addressing these challenges will be key to promoting the healthy development of the AI Agent track.

• Data Privacy and Security Issues: The effectiveness of AI Agents largely depends on their ability to acquire and analyze data. In a decentralized environment, ensuring data privacy, guaranteeing data credibility, and preventing data leakage and misuse have become difficult points in the development of AI Agent technology. The transparency characteristic of blockchain means that all data on the chain can be publicly accessed, which to some extent affects the privacy protection of user data. In AI Agent applications, especially in scenarios involving financial and personal sensitive information, how to effectively conduct intelligent learning and decision-making while ensuring data privacy has become a challenge that developers and project parties must face.

• Difficulty in Autonomous Learning and Optimization: The core value of AI Agents lies in their ability to learn autonomously and optimize decision-making processes. However, how to enable AI Agents to continuously learn and adapt in a decentralized environment is an urgent technical issue. Most existing AI models rely on centralized data sources and computing resources, which do not fully align with the characteristics of decentralized networks. Furthermore, the self-optimization capability of AI Agents is closely related to the quality of their training models and datasets. In decentralized platforms, the sources, quality, and timeliness of data may vary, leading to many uncertainties during the learning process of AI Agents.

• Technical Development and Standardization Issues: AI Agent technology is still in a rapid development stage, and there is a lack of unified technical standards and development frameworks within the industry. For instance, different AI Agent projects adopt various development tools, smart contract protocols, and data storage methods, making cross-platform applications and interoperability more challenging. To achieve more efficient development and application, AI Agent projects need to reach more consensus on technical development and promote the establishment of industry standards to facilitate the collaborative development of the entire ecosystem.

• Investment Bubble Risk: As an emerging crypto track, AI Agents have attracted a large influx of capital. However, due to the technology not being fully mature and the intense market competition, many tokens are merely meme tokens under the guise of the AI Agent concept without practical applications, potentially facing the risk of investment bubbles.

• Regulatory Uncertainty and Compliance Risks: Since AI Agents may touch on sensitive issues such as privacy protection and legal compliance concerning personal data and automated decision-making, conducting business within a regulatory framework will be an important challenge for AI Agent projects. Currently, different countries and regions have varying regulatory policies for the crypto industry, which exposes cross-border AI Agent projects to different degrees of compliance risks. For example, the EU's GDPR (General Data Protection Regulation) has strict requirements for data privacy and protection, while the regulatory environment in countries like the United States may impose higher demands on cryptocurrencies and smart contracts. To ensure the legality and sustainability of projects, AI Agent projects need to plan for compliance in advance and understand and adhere to relevant regulations.

# Trends and Summary of the AI Agent Track

The deep integration of AI Agents with the Web3 ecosystem will become another important trend in future development. Web3, as a new generation of internet architecture, emphasizes decentralization, privacy protection, and user autonomy, while AI Agents, with their characteristics of autonomous learning, intelligent decision-making, and self-optimization, can play a crucial role in the Web3 ecosystem. In the Web3 ecosystem, AI Agents can be widely applied in content creation, decentralized finance, NFT markets, DAO governance, and other fields. For example, AI Agents can provide intelligent investment advice and automated trading strategies for decentralized finance platforms by analyzing data on the blockchain. AI Agents can also help automate decision-making processes in DAOs through decentralized community governance mechanisms, reducing human intervention and improving governance efficiency.

Moreover, AI Agents will support digital identity and data privacy protection in the Web3 ecosystem. In decentralized identity (DID) systems, AI Agents can help users manage and protect their identity information and provide data analysis and intelligent recommendation services in decentralized data markets. AI Agent projects will increasingly focus on community-driven models, with project governance and decision-making becoming more decentralized. Community members will not only be users of AI Agent technology but also promoters and decision-makers in project development. The token economics of AI Agents will be closely linked to community needs, with token issuance and distribution adjusted based on community participation and contributions.

In summary, AI Agents are becoming an important component of the crypto economy, gradually influencing the ecological structure of multiple industries. From content creation and process automation to data analysis and prediction, AI Agents are ubiquitous, driving the intelligent transformation of decentralized applications and the digital economy. Although AI Agent technology still faces many challenges in its development process, as the technology matures and the market continues to expand, the application scenarios of AI Agents will continually enrich, becoming an important force driving the digital economy and Web3 development. AI Agent tokens, as the core of the AI Agent project ecosystem, will gradually form a more mature economic system, promoting more innovation and capital influx, and pushing the AI Agent track toward a more diversified and sustainable direction.

Looking ahead, AI Agents will become an important part of the crypto economy, driving the intelligent transformation of society and the economy, leading to a more efficient, fair, and innovative digital economic system. With continuous technological advancements and the gradual release of market demand, AI Agents are bound to ignite a new technological revolution and economic transformation globally.