AI Layer1 Research Report: Exploring the Fertile Ground of Decentralization in AI

Overview

In recent years, leading tech companies such as OpenAI, Anthropic, Google, and Meta have been continuously driving the rapid development of large language models (LLM). LLMs have demonstrated unprecedented capabilities across various industries, greatly expanding the realm of human imagination, and even showing the potential to replace human labor in certain scenarios. However, the core of these technologies is firmly controlled by a few centralized tech giants. With substantial capital and control over expensive computing resources, these companies have established insurmountable barriers, making it difficult for the vast majority of developers and innovation teams to compete with them.

At the same time, in the early stages of the rapid evolution of AI, public opinion often focuses on the breakthroughs and conveniences brought by technology, while the attention to core issues such as privacy protection, transparency, and security is relatively insufficient. In the long run, these issues will profoundly affect the healthy development of the AI industry and societal acceptance. If not properly addressed, the controversy over whether AI will be "good" or "evil" will become increasingly prominent, and centralized giants, driven by profit motives, often lack sufficient motivation to proactively address these challenges.

Blockchain technology, with its characteristics of Decentralization, transparency, and resistance to censorship, provides new possibilities for the sustainable development of the AI industry. Currently, numerous "Web3 AI" applications have emerged on mainstream blockchains such as Solana and Base. However, in-depth analysis reveals that these projects still face many issues: on one hand, the degree of Decentralization is limited, as key links and infrastructure still rely on centralized cloud services, and the meme attributes are too heavy, making it difficult to support a truly open ecosystem; on the other hand, compared to AI products in the Web2 world, on-chain AI still shows limitations in model capabilities, data utilization, and application scenarios, with room for improvement in both the depth and breadth of innovation.

To truly realize the vision of Decentralization AI, enabling blockchain to securely, efficiently, and democratically support large-scale AI applications, and to compete in performance with centralized solutions, we need to design a Layer 1 blockchain tailored specifically for AI. This will provide a solid foundation for open innovation in AI, democratic governance, and data security, promoting the prosperous development of a Decentralization AI ecosystem.

Core Features of AI Layer 1

AI Layer 1, as a blockchain specifically tailored for AI applications, has its underlying architecture and performance design closely aligned with the demands of AI tasks, aiming to efficiently support the sustainable development and prosperity of the on-chain AI ecosystem. Specifically, AI Layer 1 should possess the following core capabilities:

1. Efficient Incentives and Decentralization Consensus Mechanism The core of AI Layer 1 lies in building an open shared network of computing power, storage, and other resources. Unlike traditional blockchain nodes that mainly focus on ledger bookkeeping, the nodes in AI Layer 1 need to undertake more complex tasks. They must not only provide computing power and complete AI model training and inference but also contribute diversified resources such as storage, data, and bandwidth, thereby breaking the monopoly of centralized giants in AI infrastructure. This raises higher requirements for the underlying consensus and incentive mechanisms: AI Layer 1 must be able to accurately assess, incentivize, and verify the actual contributions of nodes in tasks such as AI inference and training to achieve network security and efficient resource allocation. Only in this way can the stability and prosperity of the network be ensured and the overall computing power cost effectively reduced.

2. Excellent high performance and heterogeneous task support capabilities AI tasks, especially the training and inference of LLM, have extremely high demands on computing performance and parallel processing capabilities. Furthermore, the on-chain AI ecosystem often needs to support diverse and heterogeneous task types, including different model structures, data processing, inference, storage, and other diverse scenarios. AI Layer 1 must deeply optimize the underlying architecture for high throughput, low latency, and elastic parallel requirements, and preset native support capabilities for heterogeneous computing resources to ensure that various AI tasks can run efficiently, achieving a smooth transition from "single-type tasks" to "complex diverse ecosystems."

3. Verifiability and Trustworthy Output Guarantee AI Layer 1 not only needs to prevent malicious model behavior, data tampering, and other security risks, but must also ensure the verifiability and alignment of AI output results from the underlying mechanisms. By integrating cutting-edge technologies such as Trusted Execution Environment (TEE), Zero-Knowledge Proof (ZK), and Multi-Party Computation (MPC), the platform enables independent verification of each model inference, training, and data processing process, ensuring the fairness and transparency of the AI system. At the same time, this verifiability helps users clarify the logic and basis of AI output, achieving "what is received is what is desired," enhancing user trust and satisfaction with AI products.

4. Data Privacy Protection AI applications often involve sensitive user data, particularly in the fields of finance, healthcare, and social networking, where data privacy protection is especially critical. AI Layer 1 should ensure verifiability while employing encryption-based data processing technologies, privacy computing protocols, and data permission management measures to guarantee the security of data throughout the entire process of inference, training, and storage, effectively preventing data leakage and misuse, and alleviating users' concerns regarding data security.

5. Strong ecological support and development capabilities As an AI-native Layer 1 infrastructure, the platform not only needs to have technical leadership but also must provide comprehensive development tools, integrated SDKs, operational support, and incentive mechanisms for ecological participants such as developers, node operators, and AI service providers. By continuously optimizing platform usability and developer experience, we promote the implementation of diverse AI-native applications, achieving the sustained prosperity of a Decentralization AI ecosystem.

Based on the above background and expectations, this article will provide a detailed introduction to six representative AI Layer 1 projects, including Sentient, Sahara AI, Ritual, Gensyn, Bittensor, and 0G, systematically sorting out the latest developments in the field, analyzing the current state of project development, and discussing future trends.

Sentient: Building Loyal Open Source Decentralization AI Models

Project Overview

Sentient is an open-source protocol platform that is building an AI Layer 1 blockchain (. The initial phase is Layer 2, after which it will migrate to Layer 1). By combining AI Pipeline and blockchain technology, it aims to construct a decentralized artificial intelligence economy. Its core objective is to address issues of model ownership, invocation tracking, and value distribution in the centralized LLM market through the "OML" framework (Open, Monetizable, Loyal), enabling AI models to realize on-chain ownership structure, invocation transparency, and value sharing. Sentient's vision is to empower anyone to build, collaborate, own, and monetize AI products, thereby promoting a fair and open AI Agent network ecosystem.

The Sentient Foundation team brings together top academic experts, blockchain entrepreneurs, and engineers from around the world, dedicated to building a community-driven, open-source, and verifiable AGI platform. Core members include Princeton University professor Pramod Viswanath and Indian Institute of Science professor Himanshu Tyagi, who are respectively responsible for AI safety and privacy protection, while Polygon co-founder Sandeep Nailwal leads the blockchain strategy and ecological layout. Team members have backgrounds from well-known companies such as Meta, Coinbase, and Polygon, as well as top universities like Princeton University and the Indian Institute of Technology, covering fields such as AI/ML, NLP, and computer vision, working together to promote the project's implementation.

As a second venture project of Polygon co-founder Sandeep Nailwal, Sentient was born with an aura, possessing rich resources, connections, and market awareness, which provided strong backing for the project's development. In mid-2024, Sentient completed a seed round financing of $85 million, led by Founders Fund, Pantera, and Framework Ventures, with other investment institutions including dozens of well-known VCs such as Delphi, Hashkey, and Spartan.

Design Architecture and Application Layer

Infrastructure Layer

Core Architecture

The core architecture of Sentient consists of two parts: AI Pipeline and Decentralization system.

The AI pipeline is the foundation for developing and training "Loyal AI" artifacts, consisting of two core processes:

• Data Curation: A community-driven data selection process used for model alignment.
• Loyalty Training: Ensure that the model maintains a training process consistent with community intentions.

The blockchain system provides transparency and Decentralization control for protocols, ensuring the ownership, usage tracking, revenue distribution, and fair governance of AI artifacts. The specific architecture is divided into four layers:

• Storage Layer: Stores model weights and fingerprint registration information;
• Distribution Layer: The authorization contract controls the model call entry;
• Access Layer: Verifies whether the user is authorized through permission proof;
• Incentive Layer: The revenue routing contract will allocate payment for each call to trainers, deployers, and validators.

OML Model Framework

The OML framework (Open, Monetizable, Loyal) proposed by Sentient is a core concept aimed at providing clear ownership protection and economic incentives for open-source AI models. By combining on-chain technology and AI-native cryptography, it has the following characteristics:

• Openness: The model must be open source, with transparent code and data structures, facilitating community reproduction, auditing, and improvement.
• Monetization: Each model invocation triggers a revenue stream, and the on-chain contract distributes the revenue to trainers, deployers, and validators.
• Loyalty: The model belongs to the contributor community, and the direction of upgrades and governance is determined by the DAO, with usage and modifications controlled by cryptographic mechanisms.

AI-native Cryptography

AI-native encryption utilizes the continuity of AI models, low-dimensional manifold structures, and the differentiable characteristics of models to develop a "verifiable but non-removable" lightweight security mechanism. Its core technology is:

• Fingerprint embedding: During training, a set of concealed query-response key-value pairs is inserted to form a unique signature of the model;
• Ownership Verification Protocol: Verify whether fingerprints are retained through third-party detectors (Prover) in the form of query questioning;
• License Call Mechanism: Before calling, it is necessary to obtain the "permission certificate" issued by the model owner, and the system will authorize the model to decode the input and return the accurate answer based on this.

This method enables "behavior-based authorization calls + ownership verification" without incurring re-encryption costs.

Model Rights Confirmation and Security Execution Framework

Sentient currently adopts Melange mixed security: combining fingerprint rights confirmation, TEE execution, and on-chain contract profit distribution. Among them, the fingerprint method is implemented with OML 1.0 as the main line, emphasizing the "Optimistic Security" idea, which assumes compliance by default and can detect and punish violations.

The fingerprint mechanism is a key implementation of OML. It generates a unique signature during the training phase by embedding specific "question-answer" pairs. Through these signatures, the model owner can verify ownership and prevent unauthorized duplication and commercialization. This mechanism not only protects the rights of model developers but also provides a traceable on-chain record of the model's usage behavior.

Additionally, Sentient has launched the Enclave TEE computing framework, utilizing trusted execution environments (such as AWS Nitro Enclaves) to ensure that the model only responds to authorized requests, preventing unauthorized access and use. Although TEE relies on hardware and has certain security risks, its high performance and real-time advantages make it a core component of current model deployment.