Web3-AI Landscape Report: Technical Logic, Scenario Applications, and Top Projects Depth Analysis

With the continuous rise of AI narrative, more and more attention is focused on this track. This article provides an in-depth analysis of the technical logic, application scenarios, and representative projects in the Web3-AI track, presenting a comprehensive view of the landscape and development trends in this field.

1. Web3-AI: Analysis of Technical Logic and Emerging Market Opportunities

1.1 The Integration Logic of Web3 and AI: How to Define the Web-AI Track

In the past year, AI narratives have been exceptionally popular in the Web3 industry, with AI projects emerging like mushrooms after rain. Although many projects involve AI technology, some projects only use AI in certain parts of their products, and the underlying token economics have no substantial connection to AI products. Therefore, such projects are not included in the discussion of Web3-AI projects in this article.

The focus of this article is on projects that use blockchain to solve production relationship issues and AI to address productivity problems. These projects themselves provide AI products while utilizing Web3 economic models as tools for production relationships, complementing each other. We categorize these projects as the Web3-AI track. To help readers better understand the Web3-AI track, this article will elaborate on the development process and challenges of AI, as well as how the combination of Web3 and AI perfectly resolves issues and creates new application scenarios.

1.2 The Development Process and Challenges of AI: From Data Collection to Model Inference

AI technology is a technique that enables computers to simulate, extend, and enhance human intelligence. It allows computers to perform various complex tasks, from language translation, image classification to applications such as facial recognition and autonomous driving. AI is changing the way we live and work.

The process of developing artificial intelligence models usually involves the following key steps: data collection and data preprocessing, model selection and tuning, model training and inference. For a simple example, to develop a model for classifying images of cats and dogs, you need to:

1. Data collection and data preprocessing: Collect an image dataset containing cats and dogs, which can be done using public datasets or by collecting real data yourself. Then label each image with the category (cat or dog), ensuring the labels are accurate. Convert the images into a format that the model can recognize, and divide the dataset into training, validation, and test sets.

2. Model Selection and Tuning: Choose an appropriate model, such as Convolutional Neural Networks (CNN), which is particularly suitable for image classification tasks. Adjust model parameters or architecture based on different requirements; generally speaking, the network depth of the model can be adjusted according to the complexity of the AI task. In this simple classification example, a shallower network depth may be sufficient.

3. Model Training: You can use GPU, TPU, or high-performance computing clusters to train the model, and the training time is affected by the complexity of the model and the computing power.

4. Model Inference: The files of a well-trained model are typically referred to as model weights. The inference process refers to the use of the already trained model to make predictions or classifications on new data. During this process, a test set or new data can be used to evaluate the classification performance of the model, usually assessed using metrics such as accuracy, recall, and F1-score to evaluate the model's effectiveness.

After data collection and preprocessing, model selection and tuning, and training, the trained model will be used for inference on the test set to obtain the predicted values P (probability) for cats and dogs, which is the probability that the model infers is a cat or a dog.

Trained AI models can further be integrated into various applications to perform different tasks. In this example, a cat and dog classification AI model can be integrated into a mobile application, where users upload pictures of cats or dogs to receive classification results.

However, the centralized AI development process has some issues in the following scenarios:

User Privacy: In centralized scenarios, the development process of AI is often opaque. User data may be stolen and used for AI training without their knowledge.

Data source acquisition: Small teams or individuals may face limitations on data not being open source when acquiring data in specific fields (such as medical data).

Model selection and tuning: It is difficult for small teams to access specific domain model resources or spend a lot of costs on model tuning.

Power Acquisition: For individual developers and small teams, the high purchase cost of GPUs and cloud computing rental fees can pose a significant financial burden.

AI Asset Income: Data annotation workers often cannot obtain income that matches their efforts, and the research results of AI developers are also difficult to match with buyers in need.

The challenges existing in centralized AI scenarios can be addressed by integrating with Web3, which, as a new type of productive relationship, is inherently compatible with AI representing new productive forces, thereby promoting simultaneous progress in technology and production capacity.

1.3 The Synergy Between Web3 and AI: Role Transformation and Innovative Applications

The combination of Web3 and AI can enhance user sovereignty, providing users with an open AI collaboration platform that allows them to transition from being AI users in the Web2 era to participants, creating AI that everyone can own. At the same time, the integration of the Web3 world with AI technology can spark more innovative application scenarios and gameplay.

Based on Web3 technology, the development and application of AI will usher in a brand new collaborative economic system. People's data privacy can be protected, the data crowdsourcing model promotes the advancement of AI models, and numerous open-source AI resources are available for users to utilize, while shared computing power can be accessed at a lower cost. With the help of decentralized collaborative crowdsourcing mechanisms and an open AI market, a fair income distribution system can be realized, thereby incentivizing more people to advance AI technology.

In the Web3 scenario, AI can have a positive impact across multiple tracks. For example, AI models can be integrated into smart contracts to enhance work efficiency in various application scenarios, such as market analysis, security detection, social clustering, and more. Generative AI not only allows users to experience the role of an "artist," such as creating their own NFTs using AI technology, but also creates rich and diverse gaming scenarios and interesting interactive experiences in GameFi. The rich infrastructure provides a smooth development experience, allowing both AI experts and newcomers looking to enter the AI field to find suitable entry points in this world.

2. Web3-AI Ecological Project Landscape and Architecture Interpretation

We mainly researched 41 projects in the Web3-AI track and classified these projects into different tiers. The classification logic for each tier is shown in the figure below, which includes the infrastructure layer, intermediate layer, and application layer, with each layer further divided into different sectors. In the next chapter, we will conduct a depth analysis of some representative projects.

The infrastructure layer encompasses the computing resources and technical architecture that support the entire AI lifecycle, the middle layer includes data management, model development, and validation inference services that connect the infrastructure with applications, while the application layer focuses on various applications and solutions that are directly user-facing.

Infrastructure Layer:

The infrastructure layer is the foundation of the AI lifecycle. This article classifies computing power, AI Chain, and development platforms as part of the infrastructure layer. It is the support of these infrastructures that enables the training and inference of AI models, presenting powerful and practical AI applications to users.

• Decentralized computing network: It can provide distributed computing power for AI model training, ensuring efficient and economical utilization of computing resources. Some projects offer decentralized computing power markets where users can rent computing power at low costs or share computing power to earn profits, represented by projects such as IO.NET and Hyperbolic. In addition, some projects have derived new gameplay, such as Compute Labs, which proposed a tokenization protocol, allowing users to participate in computing power leasing in different ways by purchasing NFTs that represent GPU entities to earn profits.

• AI Chain: Utilizing blockchain as the foundation for the AI lifecycle to achieve seamless interaction of AI resources on and off the chain, promoting the development of industry ecosystems. The decentralized AI market on the chain can trade AI assets such as data, models, agents, etc., and provide AI development frameworks and supporting development tools, represented by projects like Sahara AI. AI Chain can also promote technological advancements in AI across different fields, such as Bittensor, which fosters competition among different types of AI subnets through an innovative subnet incentive mechanism.

• Development Platform: Some projects offer AI agent development platforms that can also facilitate trading of AI agents, such as Fetch.ai and ChainML. One-stop tools help developers to more conveniently create, train, and deploy AI models, represented by projects like Nimble. This infrastructure promotes the widespread application of AI technology in the Web3 ecosystem.

Middle layer:

This layer involves AI data, models, as well as reasoning and validation, and can achieve higher work efficiency using Web3 technology.

• Data: The quality and quantity of data are key factors affecting the effectiveness of model training. In the Web3 world, crowdsourced data and collaborative data processing can optimize resource utilization and reduce data costs. Users can have autonomy over their data and sell their own data under privacy protection to avoid having their data stolen and profited from by malicious merchants. For data demanders, these platforms offer a wide range of choices and very low costs. Representative projects include Grass, which utilizes user bandwidth to scrape web data, and xData, which collects media information through user-friendly plugins and supports users in uploading tweet information.

In addition, some platforms allow domain experts or ordinary users to perform data preprocessing tasks, such as image labeling and data classification. These tasks may require specialized knowledge for financial and legal data processing. Users can tokenize their skills to achieve collaborative crowdsourcing of data preprocessing. For example, AI marketplaces like Sahara AI feature data tasks from different fields, covering multi-domain data scenarios; meanwhile, the AIT Protocol labels data through human-machine collaboration.

• Model: In the AI development process mentioned earlier, different types of requirements need to match suitable models. Common models for image tasks include CNN and GAN, while for object detection tasks, the Yolo series can be chosen. For text-related tasks, common models are RNN and Transformer, and there are also some specific or general large models. The depth of the models required for tasks of varying complexity also differs, and sometimes model tuning is necessary.

Some projects support users to provide different types of models or collaborate to train models through crowdsourcing. For example, Sentient, through its modular design, allows users to place trusted model data in the storage layer and distribution layer for model optimization. The development tools provided by Sahara AI come with advanced AI algorithms and computing frameworks, and they have the capability for collaborative training.

• Inference and validation: After training, the model generates weight files that can be used for direct classification, prediction, or other specific tasks. This process is called inference. The inference process is usually accompanied by a validation mechanism to verify whether the source of the inference model is correct and whether there are any malicious behaviors, etc. Inference in Web3 can typically be integrated into smart contracts, allowing for inference by calling the model. Common validation methods include technologies like ZKML, OPML, and TEE. Representative projects such as the ORA on-chain AI oracle (OAO) have introduced OPML as a verifiable layer for AI oracles. Their official website also mentions their research on ZKML and opp/ai (ZKML combined with OPML).

Application Layer:

This layer is mainly user-facing applications that combine AI with Web3 to create more interesting and innovative gameplay. This article mainly reviews projects in several areas, including AIGC (AI-generated content), AI agents, and data analysis.

• AIGC: Through AIGC, it can be extended to NFT, gaming and other tracks in Web3, allowing users to directly generate text, images, and audio through prompts (the keywords provided by users), and even generate content in games according to their preferences.