Decentralized Intelligence Network (DIN) is a fully peer-to-peer decentralized Federated Learning (FL) protocol for EVM-compatible blockchains, enabling secure AI collaboration and data monetization. It powers open AI and data marketplaces, where models can be trained without reliance on centralized Web2.0 data silos or third-party gatekeepers, supporting individually owned, data-driven economies.
We are recipients of a Cosmos Institute Grant: cosmos-institute.org, based in Austin, Texas and the University of Oxford.
Our infrastructure is ideal for applications that enable the monetization of user-owned data, while providing AI model trainers with a platform that supports FL training for decentralized data marketplaces, enabling trustless interactions between data owners and AI developers:
- For Data Owners & AI Developers: Securely monetize data while preserving privacy. Instead of sharing raw data, contributors exchange model updates, ensuring sovereignty and compliance.
- Decentralized Infrastructure: The network is secured by validators, eliminating reliance on centralized compute nodes or third-party intermediaries.
A core differentiator of DIN is our use of a decentralized validator network, secured via Proof of Stake and enhanced by our integration with Eigenlayer’s Actively Validated Services (AVS) stack. This allows us to bootstrap validator participation using Ethereum’s existing restaking infrastructure—bringing robust security and economic alignment from day one.
Our infrastructure is chain-agnostic within the Ethereum ecosystem, meaning DIN can be seamlessly deployed across a wide array of Ethereum L2s, rollups, and sidechains. This enhances interoperability and ensures DIN can serve any blockchain solution within the Ethereum family.
Rather than acting as a speculative asset, our native token serves as the backbone for decentralized infrastructure, while applications can transact using stablecoins.
This model supports:
- Stablecoin-based transactions for developers and users
- Predictable costs for AI training and data operations
- Interoperability with Ethereum-based DeFi systems
Much like traditional financial systems, this design promotes:
- Stable valuation
- Market exchange compatibility
- Price predictability
Whether using established stablecoins or experimenting with new value-pegged tokens, this architecture provides a reliable, non-volatile foundation for decentralized AI and data monetization.
-
💻 Self-Sovereign Data Ownership
Retain full control over your data while enabling AI training in a secure, decentralized infrastructure—no centralized oversight or data movement. -
🤖 Decentralized AI Training with Federated Learning
Utilize Federated Learning (FL) in a peer-to-peer network for privacy-preserving AI model updates. Data never leaves local storage—only encrypted model updates are shared, enabling secure collaboration between data providers and AI developers. -
💰 High-Frequency Trustless Incentives
Earn high-frequency stablecoin (e.g., RAI) rewards transparently, eliminating intermediaries and ensuring fair, direct compensation for all network participants. -
🔗 Eigenlayer Integration for Validator Bootstrapping
Our validator network leverages Eigenlayer's AVS stack, allowing us to inherit Ethereum’s security model while remaining chain-agnostic across Ethereum ecosystem chains—accelerating decentralization, compatibility, and trust from the outset.
Like Ethereum, no single entity owns the network. It is sustained by a global community and secured by decentralized validators, now further enhanced via Eigenlayer integration, ensuring long-term resilience and autonomy.
Decentralized Intelligence Network (DIN) Foundation is a non-profit organization dedicated to supporting the protocol’s evolution through open governance and the promotion of self-sovereign AI ecosystems.
DIN was recognized at the Summit on Responsible Decentralized Intelligence, hosted by Berkeley RDI, as a groundbreaking infrastructure for self-sovereign AI and data ownership.
📘 Read the full white paper:
Decentralized Intelligence Network (DIN) White Paper
DIN was also featured in the Oxford CS Department’s discussion on the future role of decentralized AI.
Join us in building the future of self-sovereign AI and decentralized data economies:
- 📥 Contribute: Shape the protocol — Contribution Guidelines
- 🌐 Engage: Learn more on our website
- 📧 Collaborate: Contact us at abrahamnash@protonmail.com
© 2025 Decentralized Intelligence Network (DIN)
Decentralizing AI, enabling self-sovereign data, and fostering trustless innovation.