Hetu Protocol
= "coordinating AI for Science. It’s visioned to be next generation of social contract platform". [1]
URL = https://hetu.org/
Description
"Hetu is intersubjective attribution layer, coordinating AI for Science. It’s visioned to be next generation of social contract platform, bringing humanic intelligence to Ethereum."
Characteristics
Intersubjective Consensus
• Unlike traditional blockchains that rely on objective consensus, Hetu introduces "intersubjective consensus," which allows network participants to reach agreement based on shared subjective recognition. This mechanism is applicable to areas like data value, machine learning model performance, and other domains that cannot be objectively measured using traditional methods, making it ideal for AI and scientific collaboration scenarios.
Proof of Causality Work (POCW)
• With the "Proof of Causality Work" mechanism, Hetu tracks and records the causal relationships within the system, ensuring that each network node's contributions are fairly calculated and recorded. This causality-based consensus approach enhances scalability and real-time performance, while also providing an effective incentive and reward mechanism for decentralized science and AI scenarios.
Causality Graph and Verifiable Logical Clocks (VLC)
• Hetu uses a causality graph (DAG) to establish causal ordering of events, overcoming the rigid global ordering constraints of traditional blockchains. Verifiable Logical Clocks (VLC) ensure that each node’s actions are verifiable, maintaining the integrity and fairness of causal relationships even in the presence of malicious participants.
Decentralized Value Attribution Mechanism
• Traditional blockchain mechanisms struggle with handling multi-party contributions in AI and decentralized science (DeSci). Hetu offers a transparent and trustworthy mechanism to fairly distribute rewards using causality graphs and VLC, ensuring accurate and fair attribution of all participants' contributions.
Hetu Technical Stack
• Hetu technical stack consists of a TEE-enhanced DAG chain, a causality-driven off-chain attribution layer, and a framework for AI multi-agent collaboration. By eliminating the need for traditional consensus protocols, this stack significantly boosts transaction throughput and scalability, making it highly suitable for complex AI and scientific computation tasks.
Application Scenarios
1. Decentralized AI for Science: Hetu is specifically designed for decentralized AI collaboration and scientific research. In multi-agent systems, AI agents work together to handle complex decision-making and tasks. Hetu ensures the integrity of information exchange while providing fair reward mechanisms for decentralized scientific research.
2. Decentralized Agentic Social Networks: Hetu explores how AI agents can collaborate and coordinate in decentralized networks, creating more efficient and transparent social systems.
3. Social Contracts of Future Governance: Hetu introduces social contracts, which are similar to smart contracts but focus on governance, reputation, and social consensus. These contracts are self-evolving, powered by AI agents that handle the heavy lifting. Unlike smart contracts, which are based on staking, social contracts rely on intersubjective judgment. Hetu’s governance model allocates power based on active participation, measured through non-transferable points (or keys) that cannot be bought. These points define power in governance events, making the process transparent and adaptable for each community."
More information
- Pitch Deck: https://docsend.com/view/zmp5nd7u3qwx46af774598