DaoAI
Back to research
Intelligence Infra2025.0414 min

Sentient: Building Loyal AI

Sentient 研报:构建忠诚的 AI

Read on Substack

Summary

Sentient’s OML framework (Open, Monetizable, Loyal) and model-fingerprinting approach to on-chain AI ownership — a "watermark" for large language models.

Why it matters

Open-source AGI needs more than licenses; cryptographic ownership is how contributor communities keep control.

Key ideas

01

OML = Open, Monetizable, Loyal models owned by contributors.

02

Model fingerprinting acts as a cryptographic watermark.

03

TEE + on-chain contracts enforce authorized, revenue-sharing calls.

Versions

Chinese and English versions of this report.

Chinese Article · SubstackEnglish Article · SubstackChinese Article · X

Published or syndicated by

Foresight NewsChainCatcherBiteyeHQ (X)

Related research

2025.12 · Intelligence Infra

Reinforcement Learning: The Paradigm Shift of Decentralized AI

2025.11 · Intelligence Infra

The Convergent Evolution of Automation, AI, and Web3 in the Robotics Industry