AI and web3 are the two carriages that are recognized in recent years as the next technological growth point for mankind. After ChatGPT brought a disruptive AI experience, on-chain AI has quietly completed a gorgeous transformation from meme to infrastructure, becoming the most certain new track for sustainable explosion in the minds of web3 people.
At the just-concluded Hong Kong Consensus Conference 2025, the integration of AI and web3 was also a hot topic. Whether in the main venue or the branch venue, discussions about AI x web3 were endless. The Biteye team went deep into it and learned about many potential AI projects, bringing you front-line sharing.
I. AI Infrastructure
1. AI Agent Launch Platform and Framework
With the popularity of $AI16Z and $VIRTUAL at the beginning of the year, the construction of AI infrastructure such as AI Agent launch platform and framework has been very popular in the past six months. These projects provide developers and ordinary users with a low-threshold platform for owning and using AI Agents, which is one of the key directions of this round of AI projects.
0G Labs is the first decentralized artificial intelligence operating system (deAIOS). By building an AI-specific Layer 1, connecting computing resources, data and models, it builds a distributed AI development ecosystem. Developers can use the modular tools provided by 0G to build more efficient and transparent AI Agents, while obtaining fair reward distribution. Users can contribute computing resources to the decentralized network and earn income by running nodes.
DeAgentAI is an innovative platform focused on decentralized AI Agents and is committed to promoting the development of multi-agent systems. With DeAgentAI, users can create, manage, and coordinate AI Agent networks for various scenarios such as business automation, data analysis, personalized recommendations, etc.
Autonomys Network is a decentralized infrastructure stack designed to enable secure and autonomous human-machine collaboration. Anyone can join the network with just a solid-state drive (SSD) to become part of a node. Users can create their own AI agents that can act autonomously, such as booking services, managing funds, coordinating meetings, and even participating in on-chain governance.
Gaia Network is a decentralized AI infrastructure platform that supports the distributed development and operation of AI Agents and applications. It integrates distributed storage, computing, and data verification through blockchain technology to solve the privacy, scalability, and accessibility issues of AI. Instead of building a centralized server, GaiaNet builds a distributed network of edge computing nodes controlled by individuals and enterprises to host fine-tuned AI models based on the node operators' proprietary domain knowledge and expertise.
Questflow is a decentralized network of multiple AI agents. Users only need to describe their needs, and the AI agent network can complete the task autonomously. Multiple AI agents can perform tasks seamlessly and leverage the collective intelligence and capabilities of these AI agents to achieve efficiency and speed that a single AI model or human cannot match.
2. Decentralized AI
Decentralized AI is the ultimate goal of on-chain AI. Currently, many projects are working hard for this goal in terms of computing power, data, and models, hoping to break the monopoly of large companies on LLM through decentralization and help the public obtain ownership of data and models.
Vana is committed to building a decentralized user data sovereignty platform to turn personal data into financial assets. Vana is an EVM-compatible Layer 1 blockchain that allows users to contribute and verify data through the blockchain through its unique design Data Liquidity Pools (DLP), generating tokenized assets for AI model training, while ensuring data security and quality through Proof of Contribution and privacy protection technologies. Yesterday, it received investment from YZi Lab, with CZ serving as an advisor.
Hyperbolic is an open-access AI cloud platform that provides users with affordable, scalable GPU resources and AI services by integrating global computing resources. The platform supports AI inference services, on-demand GPU access, and monetization of idle hardware, helping companies, researchers, and individuals reduce costs and efficiently use computing resources. Hyperbolic provides services for well-known on-chain AI projects such as Virtuals and AI16z, and revealed in this conference that it will launch its own AI Agent framework to further promote the innovation and application of AI computing.
OpenLedger is a next-generation network focusing on AI and blockchain, providing decentralized economic infrastructure to support developers to obtain high-quality data, fine-tune the dedicated language model (SLM) and deploy it as a payable service. SLM is its core advantage, designed for fields such as blockchain, DeFi and smart contracts; at the same time, it uses the Open LoRA framework to efficiently run thousands of fine-tuned models on a single GPU, reducing costs and maintaining high throughput. Combined with Datanets data network and Model Factory, OpenLedger provides a complete solution for data aggregation, model optimization and secure deployment.
IO.NET is a decentralized computing platform that provides on-demand access to GPU and CPU clusters, aiming to eliminate users' need for expensive hardware or infrastructure. Through the decentralized physical infrastructure network (IOG Network, which consists of independently operated distributed hardware nodes), users can instantly access distributed GPU clusters through IO Workers, which is suitable for scenarios such as high-performance computing and AI model training, without having to purchase expensive hardware.
Aethir is an innovative platform focused on providing distributed cloud computing infrastructure. Aethir Earth is a bare-metal GPU cloud designed for AI computing tasks, providing high-performance computing capabilities; Aethir Atmosphere is a cloud-based GPU network optimized for the gaming industry, supporting low-latency, high-quality gaming experience. Users can also add their own GPUs to the Aethir platform, become contributors and earn income.
MinionLab is a decentralized network of autonomous AI agents, called "Minions", which run on user devices and are used to mine data from the Internet in real time. Device owners can receive MINION token rewards by supporting the network and running these autonomous data mining agents. By leveraging idle computing resources, MinionLab provides a more resilient, cost-effective, and scalable way to mine data, while driving the AI industry towards decentralized and user-driven development.
GAIB is an economic layer solution dedicated to the field of AI and high-performance computing, aiming to create a new asset class and economic system by financializing and tokenizing GPU resources. Its core concept is to treat GPUs as assets and computing power as currency. GAIB solves the current difficulties in computing resource allocation and financing through decentralized computing networks, split ownership, and blockchain technology (L1/L2), while meeting the huge demand for high-performance computing resources in the AI era.
Kite AI is a decentralized Layer 1 blockchain platform designed for the AI economy, aiming to unlock fair access and rewards to AI assets (data, models, and agents) through its innovative Proof of AI (PoAI) consensus mechanism. Kite AI is committed to building an open and collaborative ecosystem where data providers, AI developers, and end users can benefit from the decentralized AI economy.
Automata aims to provide middle-layer privacy protection and non-tracking computing capabilities for decentralized applications (DApps). It provides developers and users with an efficient, trusted and decentralized environment to maximize data privacy and user autonomy.
Public AI's vision is to create an open and transparent AI data platform that supports the collection and annotation of multimodal data, including text, audio, video, and map data. At the same time, it provides efficient and low-cost AI-assisted data annotation services. The AI assistant will initially annotate the data, which will then be fully reviewed by the AI verifier to ensure data quality. The fair and transparent Proof of AI consensus mechanism allows everyone to participate in the AI training process by contributing data.
3. Verifiable AI
One of the important challenges facing the development of AI is the opacity of the training process and the inability to guarantee the accuracy of AI output results. Currently, many projects hope to achieve the verifiability of the AI training process and ensure the reliability of AI output results through technologies such as ZKP and TEE.
Phala Network is a decentralized cloud computing platform that provides trusted privacy computing and AI reasoning services for on-chain applications. Its products and services include a confidential computing network based on a trusted execution environment (TEE), supporting smart contracts, AI model training and reasoning, and data privacy protection. Phala Network has a natural advantage in scenarios that require high privacy, such as DeFi, DeSci, NFT markets, and on-chain AI Agent development, and can ensure data security and computational verifiability through TEE technology.
Brevis is a decentralized computing engine that aims to provide verifiable off-chain AI and blockchain computing, combining zero-knowledge proof (ZKP) to improve privacy and efficiency, focusing on financial data privacy protection, and supporting the expansion of smart contracts and Web3 applications. Its products provide a secure computing environment to ensure the privacy of on-chain AI reasoning and transactions while maintaining verifiability, and are suitable for DeFi, prediction markets, and privacy-driven AI reasoning.
Verisense Network is an innovative platform focused on decentralized data verification and trusted AI, helping developers verify data sources and ensure the authenticity and integrity of training data; it also supports auditing and verification of AI decision-making processes, providing transparent training data and reasoning processes for artificial intelligence models, eliminating the "black box" problem, and making AI decisions more credible. It is suitable for DeFi, prediction markets, and compliance monitoring that serve AI models, and can ensure the reliability of on-chain data.
II. AI Use Cases: Potential and Expectations
Compared to the rich AI infrastructure, there are still relatively few outstanding AI actual use case projects. In addition to the familiar Twitter robot AIXBT, the speakers at this event also mentioned the Virtuals ecosystem's @BukProtocol, which can realize travel ticket booking business, NBA star commentator @HeyTracyAI, and DeFAI Agent @askjimmy_ai under the ARC framework, which provide more possibilities for the application of AI Agent.
Narra is the Gamefi AI Agent platform on Berachain, which can use Narra's AI engine to generate real-time dynamic narrative content, interact with players, promote story development, and provide personalized experience in gamification scenarios. In addition, Narra also supports the generation and interaction of AI-NFTs. These NFTs are not only digital assets, but also have intelligent features and can interact dynamically with users.
AI Travel is an AI-driven travel assistant that can help users automatically customize different travel plans through chatting; it can also help users call hotels and order suitable rooms at the best prices; and provide hotel flight and other activities through browser plug-ins. Order and price comparison services.
HeyTracyAI is an AI Agent that focuses on sports commentary in the field of basketball, participated by NBA champion Tristan Thompson, and can provide real-time analysis and predictive insights for the game.
AskJimmy is an AI Agent platform focusing on finance and trading. Its main goal is to create a decentralized multi-strategy hedge fund run autonomously by AI Agents.
Third, traditional projects are transforming to AI
As the general trend, many traditional web3 projects at this event also embraced AI and announced their respective AI transformation plans.
Old blockchains such as Sui, Near, Flow and Aptos actively participated in AI-related meetings at this conference. They said in the discussion that the emergence of AI Agent is of great help in simplifying the complex blockchain interaction process, and can introduce more users to the world of web3 and solve the interaction problems that have been plaguing the industry. The tokenization and mutual cooperation of AI Agent also let everyone see more possibilities of AI on the chain. In this regard, the old blockchains have expressed the development goal of AI first, and will fully support the development of AI from the underlying architecture, account innovation and other aspects, and encourage developers to innovate and develop AI applications on the chain through activities such as hackathons.
Eigenlayer, which used to focus on Restaking services, also said in its co-hosted AI conference that it is working hard to build a decentralized trust layer (Decentralized Trust) and provide verifiable cloud services (Verifiable Cloud) to provide on-chain proof for off-chain operations such as AI training, reasoning, and prediction, and to help the development of verifiable AI Agents.
Fourth, Challenges and the Future
As a hot topic of the Hong Kong Consensus Conference 2025, the discussion and sharing on AI and Web3 were intense and exciting. While looking forward to a beautiful blueprint, many guests also mentioned that the development of on-chain AI still faces many challenges, including insufficient model reliability, ambiguity of prompt word intent, storage and hardware limitations, privacy and security, etc. These challenges not only bring technical difficulties to the industry, but also breed huge innovation opportunities. In the long run, we are full of hope for the development of on-chain AI, and look forward to promoting the integration and prosperity of AI and Web3 through further improvement of infrastructure, innovation of use cases and community collaboration.
Alex
Brian
Weiliang
Miyuki