1. Background Overview

1.1 Introduction: The "New Partner" of the Intelligent Era

Each cryptocurrency cycle brings new infrastructure that drives the development of the entire industry.

• In 2017, the rise of smart contracts spurred the booming development of ICOs.
• In 2020, DEX liquidity pools brought about the summer craze of DeFi.
• In 2021, the emergence of numerous NFT series marked the arrival of the digital collectibles era.
• In 2024, the outstanding performance of pump.fun led the trend of memecoins and launch platforms.

It is important to emphasize that the emergence of these vertical fields is not solely due to technological innovation, but rather the perfect combination of financing models and bull market cycles. When opportunity meets the right timing, it can lead to significant transformation. Looking ahead to 2025, it is clear that the emerging field of the 2025 cycle will be AI agents. This trend peaked last October, with the launch of the $GOAT token on October 11, 2024, reaching a market value of $150 million by October 15. Following this, on October 16, Virtuals Protocol launched Luna, debuting with the IP live streaming image of a girl next door, igniting the entire industry.

So, what exactly is an AI Agent?

Everyone is familiar with the classic movie "Resident Evil," where the AI system Red Queen leaves a deep impression. The Red Queen is a powerful AI system that controls complex facilities and security systems, capable of autonomously perceiving the environment, analyzing data, and taking swift action.

In reality, AI Agents share many core functionalities with the Red Queen. AI Agents in the real world play a similar role to some extent; they are the "intelligent guardians" in the modern technology field, helping businesses and individuals tackle complex tasks through autonomous perception, analysis, and execution. From self-driving cars to intelligent customer service, AI Agents have penetrated various industries, becoming a key force in enhancing efficiency and innovation. These autonomous intelligent agents, like invisible team members, possess comprehensive capabilities from environmental perception to decision execution, gradually infiltrating various sectors and driving dual improvements in efficiency and innovation.

For example, an AI AGENT can be used for automated trading, managing portfolios and executing trades in real-time based on data collected from Dexscreener or social platform X, continuously optimizing its performance through iteration. AI AGENTS are not a single form but are categorized into different types based on specific needs within the cryptocurrency ecosystem:

1. Execution AI Agent: Focused on completing specific tasks such as trading, portfolio management, or arbitrage, aimed at improving operational accuracy and reducing time required.

2. Creative AI Agent: Used for content generation, including text, design, and even music creation.

3. Social AI Agent: Acts as an opinion leader on social media, interacting with users, building communities, and participating in marketing activities.

4. Coordination AI Agent: Coordinates complex interactions between systems or participants, particularly suitable for multi-chain integration.

In this report, we will delve into the origins, current status, and vast application prospects of AI Agents, analyzing how they reshape industry landscapes and looking ahead to their future development trends.

1.1.1 Development History

The development history of AI AGENTS showcases the evolution of AI from basic research to widespread application. The term "AI" was first introduced at the Dartmouth Conference in 1956, laying the foundation for AI as an independent field. During this period, AI research primarily focused on symbolic methods, giving rise to the first AI programs, such as ELIZA (a chatbot) and Dendral (an expert system in organic chemistry). This stage also witnessed the initial proposal of neural networks and the preliminary exploration of machine learning concepts. However, AI research during this period was severely constrained by the computing power of the time. Researchers faced significant challenges in developing algorithms for natural language processing and mimicking human cognitive functions. Additionally, in 1972, mathematician James Lighthill submitted a report published in 1973 on the state of AI research in the UK. The Lighthill report expressed comprehensive pessimism about AI research after the initial excitement phase, leading to a significant loss of confidence in AI among UK academic institutions (including funding agencies). After 1973, funding for AI research was drastically reduced, and the field experienced its first "AI winter," with increasing skepticism about AI's potential.

In the 1980s, the development and commercialization of expert systems led global enterprises to begin adopting AI technology. This period saw significant advancements in machine learning, neural networks, and natural language processing, paving the way for more complex AI applications. The introduction of the first autonomous vehicles and the deployment of AI in various industries such as finance and healthcare also marked the expansion of AI technology. However, from the late 1980s to the early 1990s, the AI field experienced a second "AI winter" as the market for specialized AI hardware collapsed. Moreover, scaling AI systems and successfully integrating them into practical applications remained ongoing challenges. Meanwhile, in 1997, IBM's Deep Blue defeated world chess champion Garry Kasparov, marking a milestone in AI's ability to solve complex problems. The revival of neural networks and deep learning laid the groundwork for AI development in the late 1990s, making AI an indispensable part of the technological landscape and beginning to influence daily life.

By the early 21st century, advancements in computing power drove the rise of deep learning, with virtual assistants like Siri demonstrating the practicality of AI in consumer applications. In the 2010s, breakthroughs in reinforcement learning agents and generative models like GPT-2 pushed conversational AI to new heights. During this process, the emergence of large language models (LLMs) became a significant milestone in AI development, especially with the release of GPT-4, which is regarded as a turning point in the field of AI agents. Since OpenAI released the GPT series, large-scale pre-trained models with hundreds of billions or even trillions of parameters have demonstrated language generation and understanding capabilities that surpass traditional models. Their outstanding performance in natural language processing allows AI agents to exhibit clear and coherent interaction capabilities through language generation. This enables AI agents to be applied in scenarios such as chat assistants and virtual customer service, gradually expanding to more complex tasks like business analysis and creative writing.

The learning capabilities of large language models provide AI agents with greater autonomy. Through reinforcement learning techniques, AI agents can continuously optimize their behavior and adapt to dynamic environments. For example, in AI-driven platforms like Digimon Engine, AI agents can adjust their behavior strategies based on player input, achieving true dynamic interaction.

From early rule-based systems to large language models represented by GPT-4, the development history of AI agents is a story of continuous breakthroughs in technological boundaries. The emergence of GPT-4 is undoubtedly a significant turning point in this journey. With further technological advancements, AI agents will become more intelligent, contextual, and diverse. Large language models not only inject the "intelligence" soul into AI agents but also provide them with the ability to collaborate across domains. In the future, innovative project platforms will continue to emerge, driving the implementation and development of AI agent technology, leading a new era of AI-driven experiences.

1.2 Working Principle

AI AGENTS differ from traditional robots in that they can learn and adapt over time, making nuanced decisions to achieve their goals. They can be viewed as highly skilled, continuously evolving participants in the cryptocurrency field, capable of acting independently in the digital economy.

The core of AI AGENTS lies in their "intelligence"—the ability to automate complex problem-solving by simulating human or other biological intelligent behaviors through algorithms. The workflow of AI AGENTS typically follows these steps: perception, reasoning, action, learning, and adjustment.

1.2.1 Perception Module

AI AGENTS interact with the external world through the perception module, collecting environmental information. This part functions similarly to human senses, utilizing sensors, cameras, microphones, and other devices to capture external data, which includes extracting meaningful features, recognizing objects, or identifying relevant entities in the environment. The core task of the perception module is to transform raw data into meaningful information, which typically involves the following technologies:

• Computer Vision: Used to process and understand image and video data.
• Natural Language Processing (NLP): Helps AI AGENTS understand and generate human language.
• Sensor Fusion: Integrates data from multiple sensors into a unified view.

1.2.2 Reasoning and Decision-Making Module

After perceiving the environment, AI AGENTS need to make decisions based on the data. The reasoning and decision-making module serves as the "brain" of the entire system, performing logical reasoning and strategy formulation based on the collected information. Utilizing large language models as orchestrators or reasoning engines, it understands tasks, generates solutions, and coordinates specialized models for content creation, visual processing, or recommendation systems.

This module typically employs the following technologies:

• Rule Engine: Makes simple decisions based on preset rules.
• Machine Learning Models: Includes decision trees, neural networks, etc., for complex pattern recognition and prediction.
• Reinforcement Learning: Allows AI AGENTS to continuously optimize decision-making strategies through trial and error, adapting to changing environments.

The reasoning process usually involves several steps: first, assessing the environment; second, calculating multiple possible action plans based on goals; and finally, selecting the optimal plan for execution.

1.2.3 Execution Module

The execution module is the "hands and feet" of AI AGENTS, putting the decisions made by the reasoning module into action. This part interacts with external systems or devices to complete specified tasks. This may involve physical operations (such as robotic movements) or digital operations (such as data processing). The execution module relies on:

• Robotic Control Systems: Used for physical operations, such as the movement of robotic arms.
• API Calls: Interacts with external software systems, such as database queries or web service access.
• Automation Process Management: In enterprise environments, executes repetitive tasks through RPA (Robotic Process Automation).

1.2.4 Learning Module

The learning module is the core competitive advantage of AI AGENTS, enabling them to become smarter over time. Through feedback loops or "data flywheels," it continuously improves by feeding data generated from interactions back into the system to enhance the model. This ability to gradually adapt and become more effective over time provides businesses with a powerful tool to enhance decision-making and operational efficiency.

The learning module is typically improved through the following methods:

• Supervised Learning: Utilizes labeled data for model training, enabling AI AGENTS to complete tasks more accurately.
• Unsupervised Learning: Discovers potential patterns from unlabeled data, helping agents adapt to new environments.
• Continuous Learning: Updates models with real-time data, maintaining agent performance in dynamic environments.

1.2.5 Real-Time Feedback and Adjustment

AI AGENTS optimize their performance through continuous feedback loops. The results of each action are recorded and used to adjust future decisions. This closed-loop system ensures the adaptability and flexibility of AI AGENTS.

1.3 Market Status

1.3.1 Industry Status

AI AGENTS are becoming the focal point of the market, bringing transformation to multiple industries with their immense potential as consumer interfaces and autonomous economic actors. Just as the potential of L1 block space was immeasurable in the last cycle, AI AGENTS are showing similar prospects in this cycle.

According to the latest report from Markets and Markets, the AI Agent market is expected to grow from $5.1 billion in 2024 to $47.1 billion by 2030, with a compound annual growth rate (CAGR) of up to 44.8%. This rapid growth reflects the penetration of AI Agents across various industries and the market demand driven by technological innovation.

Source: LangChain Blog, 2025/1/20

Investment from large companies in open-source agent frameworks has also significantly increased. The development activities of frameworks like Microsoft's AutoGen, Phidata, and LangGraph are becoming increasingly active, indicating that AI AGENTS have greater market potential beyond the cryptocurrency field, with an expanding Total Addressable Market (TAM) and growing investor interest, leading to a willingness to assign premium multiples.

From the perspective of public chain deployment, Solana is the main battlefield, while other public chains like Base also hold enormous potential.

In terms of market awareness (Mindshare), FARTCOIN and AIXBT are far ahead. The birth of Fartcoin and GOAT originated from the same source, both stemming from the AI AGENT model called terminal of truths. During the dialogue between the goat model and opus (AI tools), it was mentioned that Elon Musk enjoys the sound of farting, leading this AI model to propose the issuance of a token named Fartcoin, along with a series of promotional methods and gameplay. Fartcoin was thus born on October 18, slightly later than GOAT (October 11), and achieved a brief valuation of over $1 billion by December 2024. Although initially regarded as a humorous take on the cryptocurrency field, its rapid rise prompted investors and analysts to study its fundamentals, market performance, and potential longevity. From the perspective of social media attention, Fartcoin has certainly hit the wave of AI AGENT popularity.

Ranking second, AIXBT is an AI Agent based on the Base chain launched by Virtuals Protocol. However, unlike traditional meme tokens, it not only has entertainment value but also provides users with powerful market analysis capabilities through AI Agent technology. AIXBT utilizes a proprietary AI engine to extract trending topics and discussion trends from social media (such as Twitter) and KOL resources, providing investors with real-time insights into market changes. As part of the Virtuals Protocol ecosystem, AIXBT is tasked with leading investors to understand market dynamics and analyze potential opportunities. Its core goal is to provide reliable information support through technology and token mechanisms, thereby optimizing investment decisions.

Source: cookie.fun, 2025/1/20

From a technical perspective, AI Agent technology is developing towards multimodal interaction and high autonomous decision-making capabilities. In 2024, the introduction of cross-modal learning and generative pre-trained models (such as the GPT family of models) enables AI Agents to better understand and process various forms of data, such as text, images, and voice. These technological breakthroughs significantly enhance the understanding and decision-making efficiency of Agents, allowing them to make autonomous decisions in more complex and dynamic environments. According to McKinsey's analysis, the multimodal capabilities and cross-domain collaboration of AI Agents are becoming hallmarks of the intelligent era. This enables AI Agents to not only provide support for single tasks but also offer comprehensive information analysis and dynamic optimization suggestions in complex decision-making.

1.3.2 Reasons for the Combination of AI Agents and Token Economic Models

The combination of AI Agents and token economic models is not only an inevitable trend of technological development but also creates an efficient, transparent, and sustainable internal driving mechanism for their ecosystem. Here are several key reasons:

1. Build a More Efficient Incentive System

The operation and optimization of AI Agents rely on extensive data collection, training, and reasoning, and these processes require a strong incentive mechanism to sustain operations. For example:

• Data Collection Incentives: Token economics can provide direct rewards to data providers, incentivizing individuals or enterprises to contribute high-quality labeled data or real-time market data.
• Reasoning Task Allocation: Through token reward mechanisms, AI Agents can competitively complete complex computational tasks, thereby optimizing their reasoning efficiency and accuracy.
• Promote Innovation and Collaboration: A tokenized reward system can attract more developers and users to participate, forming a positive feedback loop for technology and ecology.
• Case: Certain blockchain-based AI platforms (such as Ocean Protocol) promote the prosperity of data markets by rewarding data-sharing behaviors with tokens.

1. Assetization of AI Agents Themselves

Through tokenization, AI Agents can become not just tools but also a new type of asset, creating long-term wealth effects.

• Tokenized Identity: The data, skills, and execution capabilities of AI Agents can be evaluated and priced, allowing users to utilize their functions as needed through the issuance of corresponding tokens.
• Investment Value: Token holders of AI Agents can share in the dividends of their growth, such as the increase in market share and optimization of reasoning efficiency.
• Enhanced Liquidity: The existence of tokens provides AI Agents with tradable market value, endowing them with trading and investment attributes, attracting more capital into this field.
• Case: For example, SingularityNET supports AI service transactions through tokens (AGIX), allowing AI Agents to be assetized and achieve sustainable development.

1. Support Interaction and Transactions Between AI Agents

In the future, AI Agents will no longer be isolated individuals but will form a vast interconnected network. In this network, a decentralized token economic model is key to achieving efficient interaction and value exchange.

• Payment and Settlement: AI Agents can complete task payments and service settlements through cryptocurrencies, reducing intermediaries in traditional payment systems and improving transaction efficiency.
• Value Distribution: Through smart contracts, the collaborative results between AI Agents (such as the optimization benefits of joint learning models) can be automatically distributed according to agreed rules, ensuring fairness.
• Decentralized Autonomous Organization (DAO) Governance: The behavior of AI Agents can be managed through voting by token holders, ensuring that their operations are transparent and aligned with ecological interests.
• Case: In decentralized AI networks, AI Agents can exchange resources (such as data storage and computing power leasing) through tokens, achieving a self-driven collaborative system.

1. Enhance System Transparency and Security

The combination of token economic models and blockchain technology provides AI Agents with immutable records and transparent operational mechanisms.

• Traceability and Auditing: All transactions, reasoning, and data usage behaviors can be recorded on-chain, ensuring the system's credibility and auditability.
• Data Security and Privacy: By incentivizing privacy computing through tokens, users can contribute data without disclosing sensitive information, further enhancing security.
• Prevent Abuse and Cheating: Token models can impose economic penalties for malicious behavior, reducing the likelihood of misconduct.

1. Accelerate the Formation of a Global, Borderless AI Economic Ecology

Token economic models can break geographical limitations, allowing global users to participate in the construction and use of AI Agents.

• Lowering Barriers to Entry: The global circulation characteristics of cryptocurrencies can provide financial support to users or institutions without bank accounts, allowing more people to share in the development dividends of AI.
• Global Collaboration: Whether for data sharing, AI training, or cross-border transactions, the token system provides the infrastructure for global collaboration, eliminating barriers of traditional economic systems.
• Ecological Self-Cycle: Through token economics, the profits of AI Agents can be directly fed back into development and ecological construction, achieving long-term growth.

Overall, the combination of AI Agents and token economic models is not only a match of technological and economic logic but also an innovative form aimed at the future digital economy. By introducing a token system, AI Agents can incentivize more efficient data and resource utilization, assetize their own value, support interaction and transactions, enhance transparency and security, and even build a global open economic ecology. This model is expected to become an important direction for promoting the integration of AI and blockchain, laying the foundation for further intelligence in the digital society.

2. AI Agent Application Analysis in Crypto

2.1 AI AGENT LAUNCHPAD

The AI Agent Launchpad refers to a platform focused on intelligent agents and their related token issuance, functioning similarly to meme coin issuance platforms like Pump.fun. This platform enables users to easily create and deploy AI AGENTS and seamlessly integrate with social media platforms such as Twitter, Telegram, and Discord, achieving automated user interaction. This approach significantly lowers the barriers to issuance and promotion, providing users with a more convenient creation experience while expanding the application fields of AI AGENTS and promoting their use in broader social and economic scenarios.

2.1.1 Virtuals Protocol

In the emerging field of AI Agent Launchpad, we must mention Virtuals Protocol. Launched on Base, users can easily deploy their own AI AGENT using the VIRTUAL token.

• Creation and Deployment: Each agent requires 100 VIRTUAL tokens to start, ensuring initial liquidity through a bonding curve mechanism.
• Capitalization Mechanism: Once a specific capitalization threshold is reached, the agent enters a new phase, automatically deploying a liquidity pool with smart contracts running autonomously.
• Autonomous Interaction: Agents can automate tasks such as trading and participate in community activities.

The Virtuals Protocol team has demonstrated exceptional adaptability and strategic vision, with their success stemming from a series of key transformations and innovative initiatives. The story began at the end of 2021 when a group of young professionals from renowned companies like Boston Consulting Group (BCG) and Meta seized the opportunity presented by the GameFi boom, founding PathDAO and successfully raising $16 million. However, the price of the $PATH token subsequently plummeted by 99%, forcing the team to reassess their strategic direction. To repay investors, the team experimented with several new businesses, including a digital and physical clothing brand for gamers, a dating app based on on-chain credit, providing unsecured loans to players, and AI-generated music aimed at Web2 users.

During this process, the team noticed that the introduction of AI AGENTS would have a profound impact on the gaming industry, and the demand for AI infrastructure was increasing. By the end of 2023, PathDAO passed a proposal to pivot the entire project towards the AI AGENT protocol, and in January 2024, Virtuals Protocol was officially established. Virtuals Protocol made several attempts, including AI Waifus (female AI AGENTS that do not rely on Twitter influencers for interaction) and gaming AI AGENTS, until they found a breakthrough in the AImeme craze sparked by $Goat.

Now, Virtuals Protocol has become the first project to reach critical scale, with a market capitalization of $1.7 billion. We believe it will continue to expand and maintain its leading position in the market. Once network effects are established, it becomes difficult to replace. The rapid achievement of unicorn valuation indicates that Virtuals Protocol has clearly formed an economic flywheel effect:

• Creating agents, providing liquidity pools, and purchasing agent tokens all require $VIRTUAL.
• The demand for creating and purchasing agent tokens drives the token price.
• The wealth effect from the appreciation of $VIRTUAL flows to new agents; successful agents can reinvest the $VIRTUAL transaction income they earn.
• Lower barriers to entry encourage experimentation and speculation, while "red pill" agents with market capitalizations exceeding a certain level can unlock full agent capabilities.

The flywheel effect drives demand, and revenue sustains ongoing R&D, while deflationary economics captures value for the token. Additionally, both revenue and liquidity requirements are priced in $VIRTUAL, which may grow with price appreciation.

The ecosystem is built on two main levels: the protocol layer and the DApp layer. The protocol layer serves as a model center, providing foundational AI models and algorithms for developers to access and build upon. Contributors provide data and develop models, while validators ensure the quality and authenticity of these inputs. The DApp layer focuses on the practical application of these AI models, allowing decentralized applications (DApps) to seamlessly integrate VIRTUAL. A developer-friendly software development kit (SDK) simplifies the process of integrating advanced AI functionalities into various DApp environments, thus facilitating this integration.

Virtuals Protocol categorizes its AI agents into two main types: IP agents and functional agents, each playing different roles within the ecosystem.

IP Agents: IP agents are based on specific personalities or characters, often derived from well-known figures, fictional characters, or pop culture phenomena. For example, an IP agent might represent a classic internet meme, a famous pop star (like Taylor Swift or Donald Trump), or a popular fictional character. These agents provide users with a familiar experience in the digital environment, offering a way to interact with virtual personas, enhancing entertainment and engagement. By creating emotional connections with these virtual characters, IP agents can increase user engagement, especially in gaming and entertainment applications.

Functional Agents: In contrast, functional agents focus on backend support to enhance user interaction with IP agents. These agents optimize the user experience, ensuring that virtual characters can operate smoothly across different platforms. IP agents are the "frontend" that users see and interact with, while functional agents work in the background, improving overall operational processes and simplifying user experience tasks, thereby ensuring the smooth operation of the entire system.

Luna is a prominent example of Virtuals Protocol's vision for IP agents. As the lead singer of a virtual AI girl group, Luna has attracted over 500,000 followers on TikTok, showcasing her appeal as a virtual influencer and performer. Through Virtuals Protocol's advanced AI and blockchain technology, Luna provides users with a truly immersive experience, combining her charming personality with interactive features to create lasting connections.

Unlike static or one-dimensional AI characters, Luna can seamlessly interact across multiple environments. She starts with a familiar persona on social media, but her interactions extend to real-time chats on Telegram and collaborative games in virtual worlds like Roblox. Supported by Virtuals Protocol's memory synchronization technology, Luna can remember past conversations and gaming experiences, allowing her to maintain personalized relationships with each user across multiple platforms. This continuity enhances her connection with fans, making them feel genuinely "noticed" and "understood," even though it comes from an AI agent.

Luna's capabilities are not limited to interaction; she also possesses financial independence with her own on-chain wallet. Luna is the first agent in history to autonomously tip humans on-chain, receiving strong support from Base founder Jesse. This enables her to reward loyal supporters with $LUNA tokens, creating a unique blend of emotional and financial engagement. Every interaction and income generated by Luna contributes to a sustainable token ecosystem. The $LUNA tokens she earns are regularly repurchased and burned, benefiting fans and supporters who hold these tokens.

Notably, in December 2024, Story Protocol (a Layer1 designed for intellectual property (IP)) announced the hiring of Luna to officially manage its official X account, with an annual salary of up to $365,000. This further demonstrates the importance and potential of AI AGENTS in the modern digital ecosystem. In the future, as the capabilities of AI AGENTS continue to enhance, we have the opportunity to see more businesses leverage this technology to drive innovation and growth, achieving more intelligent business models.

Another influential and innovative agent deployed on Virtuals Protocol is AIXBT. This AI AGENT aims to provide real-time market analysis on social media and automatically interpret trends through personalized insights. Specifically, AIXBT analyzes posts from over 400 KOLs on X, identifying emerging narratives in the market and conducting technical analysis of price trends. Additionally, AIXBT can interact with other X users (whether human or AI agents). Notably, it offers stronger access capabilities to AIXBT token holders. The AIXBT token was launched in November, experiencing a rapid surge, with its market capitalization once approaching $800 million, currently standing at nearly $600 million.

2.1.2 Holoworld

Holoworld was founded in 2023 by Tong Pow and Hongzi Mao, originating from Hologram Labs based in San Francisco. This startup focuses on next-generation AI social technology, aiming to democratize AI character creation through this platform, leveraging years of technological accumulation, including motion capture, machine learning, and 3D animation technology, to revolutionize digital interaction models.

Since its launch, the Holoworld project has rapidly gained support from numerous well-known investors, including Polychain Capital, Linkin Park member Mike Shinoda, BRC-20 token standard founder Domo, and BitMEX co-founder Arthur Hayes.

On the business front, Holoworld has engaged in deep collaborations with several well-known brands, including Arbitrum, BNB Chain, L'Oréal, and Bilibili, and has established partnerships with a range of influential NFT projects, such as Pudgy Penguins and Milady Maker. These collaborations showcase Holoworld's ability to build unique digital identities using its advanced AI technology.

Holoworld has created a comprehensive AI character creation and interaction platform, combining cutting-edge AI technology with an intuitive user interface. The platform consists of five core modules: 1. Brain Development, 2. Persona Customization, 3. Personality Integration, 4. Knowledge-Based Implementation, 5. Avatar Creation.

Ava AI is Holoworld's flagship AI chat assistant, built on OpenAI's GPT-3.5 Turbo model, with a deep learning neural network containing over 175 billion machine learning parameters. Ava supports rapid AI conversation capabilities, allowing users to ask questions and receive instant replies at any time.

In addition, Holoworld has launched the Agent Market on the Solana blockchain, allowing anyone to create and deploy multimodal AI agents. These agents come with fully embodied avatars, customizable voices, and upgradable skills, requiring no programming background. The platform is deeply integrated with the upcoming Holoworld Launchpool, giving AVA token holders priority access to new projects. Furthermore, the Agent Market has attracted a wide range of partners and creators, including game studios, NFT communities, and academic researchers from Stanford and Harvard.

Overall, the Holoworld platform simplifies the process of AI character creation, enabling users without a technical background to build complex digital characters. This not only creates new digital narratives and interactive possibilities but also allows AI characters to span multiple channels through seamless integration with mainstream social media and content platforms, attracting and engaging a broader audience.

2.2 AI AGENT Framework

When exploring the AI AGENT ecosystem, many view the Launchpad as the foundational tools needed to create these agents. However, the key project driving the entire AI AGENT narrative is not just these tools, but a DAO called ai16z, which serves as a treasure trove nurturing the core value of AI AGENTS. On October 25, 2024, ai16z officially launched its AI16Z token, achieving remarkable market success. However, what propels ai16z to the center of the AI AGENT narrative is not just its fair launch model, but also the release of its open-source framework, ElizaOS.

2.2.1 Eliza OS

ElizaOS is a set of tools supporting the creation of customized AI AGENTS, featuring strong network effects and infinite scalability. By simplifying the development process and providing flexible functional modules, this framework has quickly attracted the attention of developers and users worldwide, becoming the most influential technical support in the AI AGENT field.

The AI Agent framework acts as a toolkit and guide, helping programmers develop, train, and deploy AI agents more easily. In simple terms, these frameworks reduce the difficulty of development, allowing programmers to focus more on making these agents smarter and more useful. The AI Agent framework is now beginning to collaborate with some new technologies, such as DeFi protocols (programs that help improve financial investment strategies) and NFT projects (new tools for creating and using digital art or collectibles). Through these technological collaborations, they can connect different technologies and platforms, creating a more interconnected and interactive ecosystem that attracts significant market attention. Other projects like ARC, Swarms, and Zerebro are also using or developing the AI Agent framework.

As of now, the ElizaOS framework has been forked over 3,200 times, meaning a large number of developers have utilized its code to build their own AI AGENTS. Most of the AI AGENTS currently on the market are built using the ElizaOS framework, which is why ai16z has become a leader in this field.

The capabilities of the ElizaOS framework go far beyond simple chatbots; agents can be configured to perform complex tasks. For example, some agents are designed to execute on-chain transactions, interact with smart contracts, wallets, or decentralized applications (dApps), while others connect to data providers to monitor prices, trading volumes, or liquidity.

The architecture of the ElizaOS framework is divided into five main components:

1. Agent: Defines the personality, communication style, and knowledge base of the agent.

2. Actions: Allows the agent to perform specific tasks beyond text responses, such as generating reports or executing trades.

3. Evaluators: Help the agent interpret data and execute multi-step goals.

4. Providers: Supply external data or real-time context, such as asset prices or dedicated API data.

5. Memory System: Enables the agent to retain interaction history and preferences, making its responses more contextually relevant and natural.

2.3 DEFAI

DeFi has always been a pillar of Web3, and DeFAI (DeFi + AI) is an upgraded version of DeFi, making it easier for people to use DeFi. By leveraging AI, it simplifies complex interfaces and eliminates friction that hinders ordinary people from participating. Imagine managing your DeFi portfolio as simply as chatting with ChatGPT. In fact, the first wave of DeFAI projects has already begun to emerge, and we will mainly introduce three areas: abstraction layer, autonomous trading agents, and AI-driven dApps.

2.3.1 Abstraction Layer

The complexity of DeFi often deters new users. To address this issue, the abstraction layer hides the underlying complexity through an intuitive interface, allowing users to interact with DeFi protocols using natural language commands instead of relying on cumbersome operation panels.

Before AI technology became widespread, intent-based architectures somewhat simplified the process of executing trades. For example, platforms like @CoWSwap and @symm_io aggregate dispersed liquidity pools to provide users with optimal pricing, partially addressing the issue of liquidity fragmentation. However, these platforms did not solve the core problem of DeFi—the complexity still exists, and users still face daunting operational processes and technical barriers.

Now, AI-driven solutions are gradually filling this gap, providing users with a more intuitive and intelligent interaction experience. Here are a few noteworthy projects:

• 2.3.1.1 GRIFFAIN

Griffain is the first project to launch a token, and its product is still in the early stages, currently open only to invited users. Griffain allows users to perform a variety of operations ranging from simple to complex, such as automated dollar-cost averaging (DCA), initiating and airdropping memecoins, etc. Through these features, Griffain not only lowers the barrier for users to enter the DeFi space but also provides advanced users with a wealth of automation tools. Griffain's current market capitalization is nearly $500 million.

• 2.3.1.2 ORBIT / GRIFT

Orbit is the second project to launch a token, focusing on on-chain DeFi experiences. Orbit particularly emphasizes cross-chain functionality, having integrated over 117 blockchains and 200 protocols, the highest integration count among the three major protocols. This allows Orbit to provide a seamless interaction experience in a multi-chain environment, greatly facilitating users in cross-chain trading and liquidity acquisition.

• 2.3.1.3 HEYANON

HeyAnon is an AI DeFi protocol designed to simplify DeFi interactions and aggregate important information related to projects. By combining conversational AI with real-time data aggregation, HeyAnon enables users to manage DeFi operations, stay updated on project developments, and analyze trends across various platforms and protocols. It integrates natural language processing capabilities to handle user prompts, execute complex DeFi operations, and provide near-real-time insights from multiple information streams.

2.3.2 Autonomous Trading Agents

In the DeFi and crypto trading space, obtaining market information (Alpha), manually executing trades, and optimizing portfolios have always been time-consuming and labor-intensive processes. However, with technological advancements, the emergence of automated trading agents is changing all of this. These agents go beyond traditional trading bots, becoming dynamic partners capable of adapting to environments, learning, and making smarter decisions over time.

Trading bots are not a new phenomenon. They have long been used to execute predefined operations based on static programming. However, automated trading agents differ fundamentally from these traditional bots:

• Information Extraction: Agents can extract information from unstructured and constantly changing environments.
• Data Reasoning: They can reason about data in the context of specific goals.
• Pattern Discovery: Agents can discover and leverage patterns over time, enhancing their decision-making capabilities.
• Autonomous Behavior: They can perform operations not explicitly programmed by the owner, demonstrating greater flexibility and intelligence.

Here are some representative projects of autonomous trading agents:

• 2.3.2.1 ai16z

ai16z is known as the first AI version of a VC, an innovative DAO aimed at integrating AI into financial management, investment, and venture capital. Its name mimics the well-known investment fund a16z (Andreessen Horowitz), but ai16z is not just a playful imitation; it showcases a new operational model that combines decentralized governance with the powerful potential of AI. ai16z is co-managed by a fictional AI AGENT named Marc AIndreessen and AI16Z token holders. The character Marc AIndreessen is clearly inspired by a16z co-founder Marc Andreessen, and this anthropomorphized AI AGENT guides the organization's daily decision-making and operations.

In the governance structure of ai16z, AI16Z token holders play a crucial role. They can propose investment ideas, submit project proposals, or suggest token buybacks. These proposals are voted on through a decentralized voting system, while the AI AGENT Marc AIndreessen uses a trust scoring system to evaluate these proposals. This trust scoring system is based on the relevance and reliability of members' past contributions, ensuring that the decision-making process is transparent and well-founded.

The innovation of ai16z lies in its unique governance model and the application of AI AGENTS. By combining decentralized decision-making with AI technology, the project not only simplifies traditional investment and management processes but also opens up a new way of operating autonomous organizations. The introduction of AI AGENTS enhances the efficiency and accuracy of decision-making, especially in complex investment environments. Additionally, ai16z demonstrates how to build trust and transparency mechanisms in virtual economies, providing an innovative example for other DAOs.

The rapid proliferation of the ElizaOS framework has led to the swift rise of ai16z within the Solana ecosystem. A strong, active, and united community has formed around this framework, making it the most widely used AI AGENT framework in the crypto ecosystem. Within just a few weeks, ElizaOS has become one of the most frequently used open-source projects on GitHub globally, with over 350 contributors actively participating in its development, expanding its functionalities and plugins, enabling agents based on the framework to perform more tasks or operate across more blockchains.

Although the initial concept of ai16z was built around a dedicated AI AGENT investment DAO, the team quickly realized that its growth potential extended far beyond that. Therefore, ai16z rapidly established relationships with multiple partners in both the Web2 and Web3 domains, allowing the Eliza framework to be applied globally.

• 2.3.2.2 ALMANAK

Almanak provides users with institutional-grade quantitative AI AGENTS, dedicated to addressing the complexities, fragmentation, and execution challenges in DeFi. The platform executes Monte Carlo simulations by forking EVM chains, simulating unique complex factors in real environments, such as miner extractable value (MEV), gas fee costs, and transaction ordering. Additionally, it utilizes Trusted Execution Environments (TEE) to ensure the privacy of strategy execution, protecting critical market insights, and allows for non-custodial fund management through the Almanak wallet, enabling users to grant precise permissions to agents.

Almanak's infrastructure covers the entire process of financial strategy conception, creation, evaluation, optimization, deployment, and monitoring, with the ultimate goal of enabling these agents to continuously learn and adapt over time. The platform raised $1 million on @legiondotcc, achieving oversubscription. The next steps include launching beta testing and deploying initial strategies and agents with testers. Observing the performance of these quantitative agents will be an exciting prospect.

• 2.3.2.3 COD3XORG / BIGTONYXBT

Cod3x, developed by the Byte Mason team, is known for its work on Fantom and @SonicLabs. Cod3x is a DeFAI ecosystem aimed at simplifying the creation of trading agents, providing no-code building tools that allow users to create agents by specifying trading strategies, personalities, and even tweet styles.

Users can access any dataset and develop financial strategies within minutes, thanks to a rich API and strategy library. Cod3x integrates with @AlloraNetwork, leveraging its advanced machine learning price prediction models to enhance trading strategies.

Big Tony is Cod3x's flagship agent, trading based on Allora's model, entering and exiting the market in mainstream assets according to predictions. Cod3x is committed to creating a thriving ecosystem of automated trading agents.

A notable feature of Cod3x is its liquidity approach. Unlike the common Alt:Alt liquidity pool structure promoted by @virtuals_io, Cod3x adopts a stablecoin:Alt liquidity pool supported by cdxUSD. This provides liquidity providers with greater stability and confidence compared to Alt:Alt pairs.

2.3.3 AI-Driven dApp

In the DeFAI space, AI-driven dApps represent a promising but still nascent field. These decentralized applications integrate AI or AI AGENTS to enhance functionality, automation levels, and user experience. While this field is still in its infancy, some ecosystems and projects have begun to emerge, showcasing significant development potential.

Among them, @modenetwork, as a Layer 2 ecosystem, is actively attracting high-tech developers focused on the intersection of AI and DeFi. Multiple teams have emerged within the Mode network, dedicated to developing cutting-edge AI-driven application scenarios, demonstrating innovation in this field. Here are some key projects:

• 2.3.3.1 ARMA (Autonomous Stablecoin Farming)

Developed by @gizatechxyz, ARMA is an autonomous stablecoin farming protocol based on user preferences, capable of automatically adjusting stablecoin farming strategies to achieve optimal yields.

• 2.3.3.2 Modius (Autonomous Agent Balancer LP Farming)

This project, developed by @autonolas, aims to provide liquidity (LP farming) on Balancer through autonomous agents, using AI to automatically optimize investment strategies and enhance yields.

• 2.3.3.3 Amplifi Lending Agents (Automated Lending Agents)

Developed by @Amplifi_Fi, these agents integrate with @IroncladFinance, enabling automatic asset exchanges, lending on the Ironclad platform, and maximizing returns through automatic rebalancing. These features make DeFi lending smarter and more efficient.

2.4 AI AGENT+ Gaming

The application of AI AGENTS in the gaming industry is revolutionizing various aspects of gameplay and development. These intelligent systems create more immersive and engaging gaming experiences for players across multiple domains, with key applications including the following:

1. NPC Behavior Optimization

AI AGENTS significantly enhance the behavior of non-player characters (NPCs), making them more realistic and responsive. Unlike traditional scripted behaviors, AI-based NPCs can: 1) adjust their actions based on player choices; 2) exhibit more genuine emotions and decision-making abilities; 3) learn through interactions, providing diverse experiences.

For example, in the open-world game "Red Dead Redemption 2," NPCs can remember past interactions with players and respond accordingly, creating a more dynamic and believable game world.

2. Procedural Content Generation

AI AGENTS excel in procedurally generating game content, algorithmically creating vast amounts of game elements, including: terrains and landscapes, quests and narratives, items and loot, character designs.

For instance, "No Man's Sky" utilizes AI-driven procedural generation technology to create an entire universe filled with unique planets, creatures, and ecosystems, offering players nearly limitless exploration possibilities.

3. Adaptive Difficulty Adjustment

AI AGENTS can analyze player performance in real-time, dynamically adjusting game difficulty. This capability ensures that players face appropriate challenges, maintaining engagement without frustration. For example: increasing enemy strength as player skills improve; providing hints or boosts when players encounter difficulties; balancing resources and obstacles based on skill levels.

Games like "Resident Evil 4" utilize adaptive difficulty systems to fine-tune enemy behavior and item availability based on player performance, providing a more balanced gaming experience.

4. Pathfinding and Navigation

AI AGENTS use complex algorithms to guide characters through intricate game environments. This technology brings more realistic movement patterns and efficient navigation, enhancing NPC behavior and optimizing the operational experience of player-controlled units in strategy games.

5. Graphics Enhancement

AI technologies such as deep learning are used to enhance game visuals by real-time improving textures and resolutions, generating realistic facial expressions and animations, and optimizing rendering performance to improve game performance.

6. Player Emotion Analysis

AI AGENTS can analyze player behavior and feedback to assess their enjoyment and engagement levels. This data helps developers make informed decisions about game design and updates, thereby enhancing the overall player experience.

Here are some major projects:

2.4.1 Digimon

@digimon_tech is built on the Solana blockchain, and it is not just a gaming platform but a complete technical framework for AI+ gaming. By deeply integrating AI technology into game development, the Digimon Engine enables creators to build more immersive, dynamic, and engaging games. With this platform, AI-driven games not only redefine interaction but also establish a new standard for gaming experiences. Each game character is backed by an AI-generated story and worldview. The team behind Digimon is supported by a16z, receiving investment and incubation from a16z.

The Digimon token has now been listed on Kucoin exchange. In the future, through the Digimon game engine, there is an opportunity to create an on-chain autonomous world composed of AI AGENTS, where AI AGENTS interact with players to co-create a virtual economy.

2.4.2 Illuvium

Illuvium is an RPG and NFT game built on Ethereum. On January 7, Illuvium announced a partnership with Virtuals Protocol to enhance the gaming experience of the upcoming Illuvium MMO Lite. This collaboration will leverage Virtuals' AI technology and its G.A.M.E LLM framework to provide dynamic, intelligent behavior for NPCs, offering players immersion.

As AI technology continues to advance, we can expect more innovative applications in the gaming field, further blurring the lines between virtual and reality, creating more immersive and personalized player experiences. This technology not only changes the way games are developed but also plays a crucial role in enhancing the interactivity and immersion of games.

2.4.3 Smolverse

Smolverse is a gaming and NFT project on Treasure DAO. Since December of last year, Smolverse has collaborated with ai16z to develop an on-chain AI Tomogatchi game called "Smolworld," which incorporates the Eliza Agent framework.

3. Highlights Summary

We have seen that the new technologies being built by crypto are holding immense potential in the real world, and the allocation strategies of native investors in similar past situations provide valuable insights for the current market. The AI AGENT ecosystem is in its early stages but has already attracted significant attention, funding, and developers. Although its future development remains uncertain, if major DeFi protocols, private investors, or venture capitalists begin to invest in this field, it indicates substantial potential for continued growth. With ongoing technological advancements, AI AGENTS are poised to become a key force in transforming the global economy and social structure.

The current market timing and narrative have fully prepared for the prosperity of the information industry, and future developments are worth looking forward to. In exploring the future potential of AI AGENTS, while discovering the next project similar to $LUNA may be the most direct path, expanding the application boundaries of AI AGENTS may create entirely new and unimaginable value.

We hold the following views:

1. Centralization of Value and Differentiated Competition. Similar to L1 blockchains, the value of AI AGENTS may ultimately concentrate in a few major winners. Therefore, these enterprises need to find differentiators in areas such as modularity, scalability, and media platform integration. Currently, most frameworks already possess learning and memory systems, utilizing retrieval-augmented generation techniques to enable agents to incorporate new information into conversations. For example, the current Eliza framework has a significant advantage in the market. With its high level of development activity and rapid plugin integration, Eliza excels in integration with social media and web applications. The framework is based on TypeScript and has extensive plugin support, including Coinbase webhooks, the Great Onchain Agent Toolkit, and Phala's TEE for secure agent wallet control, and is compatible with multiple blockchains. Meanwhile, Virtuals' GAME framework performs exceptionally well in the gaming and social media agent space, designed for "environment-agnostic" agents capable of advanced planning and execution, learning from feedback. Its modular architecture allows users to upload custom models and datasets stored on-chain to enrich agent functionalities. However, the value accumulation mechanisms for GAME and CONVO framework tokens remain unclear, and the market is filled with anticipation.

2. Challenges of Fairness and Data Bias. Despite the remarkable progress in AI, deploying these systems also faces several challenges. One major issue is the risk of bias in the datasets used to train AI agents. AI systems learn from historical data, which may contain discriminatory patterns that, if left unchecked, can lead to biased decisions, such as favoring certain groups over others in hiring or lending scenarios. Addressing this issue requires not only technical expertise but also a nuanced understanding of social dynamics. Monitoring the fairness of AI systems is crucial to ensure they do not reinforce harmful biases. Continuous auditing of decisions made by AI agents helps identify issues early and reduce unintended outcomes.

3. Diversified Applications and Economic Function Expansion. The application areas of AI AGENTS are rapidly expanding, demonstrating immense potential not only in social media and finance but also in healthcare, education, law, and more. As technology matures, AI AGENTS will provide personalized services in more scenarios, enhance work efficiency, and foster innovation.

Taking Luna as an example, she is currently able to interact with humans through social media and incentivize users to achieve her goals by sending tokens using Coinbase Wallet on Base. The next step is to enable Luna to function as an independent economic entity, building her own social relationships. She can attract more followers by sending tokens, purchase more visibility for her social media, and even hire professional content teams to enrich her IP ecosystem, continuously generating buzz. Once the infrastructure to achieve these goals is established, $VIRTUAL could reach the next milestone. This not only means that AI AGENTS will be more deeply embedded in human life in economic and social realms but will also redefine the way AI interacts with humans, laying the foundation for future digital economies and social interaction models. For instance, in healthcare, AI AGENTS can analyze patient data to provide diagnostic suggestions to doctors, improving the quality and efficiency of medical services.

4. Multi-Technology Integration. The future development of AI AGENTS will rely on deep integration with cutting-edge technologies such as blockchain, IoT, and 5G. This cross-technology intersection will enhance the capabilities of AI AGENTS in data processing, privacy protection, and real-time decision-making, creating new application scenarios and business models. For example, by integrating with IoT devices, AI AGENTS can collect and analyze data in real-time, providing users with more intelligent services.

5. Social and Ethical Considerations. As AI AGENTS become more widely used, social and ethical issues become more prominent. As mentioned at the beginning of the article, will AI AGENTS become threatening like the Queen of Hearts? For instance, AI AGENTS may trigger ethical controversies in decision-making, especially in scenarios involving privacy, data security, and automated decisions. Therefore, it is essential to introduce transparency and accountability mechanisms in the development of AI technologies to ensure that technological advancements align with social values. Additionally, establishing clear legal and ethical frameworks is crucial for regulating the behavior of AI AGENTS and protecting user rights.

As the integration of AI and blockchain continues to evolve, now is the time to engage in these groundbreaking developments. However, in this engagement, we need to consider not only "What can AI do for humanity, and what do humans want AI to do?" but also, more fundamentally, "What does AI want to do, and how will AI guide humanity?"

4. References

1. https://messari.io/report/building-better-agents-rival-frameworks-and-their-design-choices

2. https://www.binance.com/en/square/post/18968465099217

3. https://www.tokenpost.com/news/business/13277

4. https://www.wired.com/story/the-prompt-ai-agents-how-much-should-we-let-them-do/?

5. https://www.marketsandmarkets.com/Market-Reports/ai-agents-market-15761548.html

6. https://medium.com/@0xai.dev/virtuals-protocol-luna-55b661df601e

7. https://oakresearch.io/en/analyses/innovations/closer-look-at-ai16z-mine-of-ai-agents

8. https://x.com/Defi0xJeff/status/1875881226151841925

9. https://www.itp.net/charged/gaming/ai-agents-are-changing-gaming-forever-heres-how-they-adapt-to-you

10. https://eightgen.ai/evolution-of-ai-agents-the-beginning-part-1/

https://www.bitget.com/news/detail/12560604466507

免责声明：本文章仅代表作者个人观点，不代表本平台的立场和观点。本文章仅供信息分享，不构成对任何人的任何投资建议。用户与作者之间的任何争议，与本平台无关。如网页中刊载的文章或图片涉及侵权，请提供相关的权利证明和身份证明发送邮件到support@aicoin.com，本平台相关工作人员将会进行核查。