Skip to content

Decentralization

This document outlines process of progressive decentralization of near.ai capabilities.

near.ai project approaches decentralization in a progressive way, starting with delivering best researcher, developer and user experience possible and progressively decentralizing and enabling privacy and trust features over time.

Component Objective
Registry Decentralized storage with support of private and encrypted items
Training / Fine-tuning Leveraging decentralized set of nodes to train models in a provable way
Agent runner Using trusted execution environment to run agents privately and
Agent memory Using FHE to do retrieval from encrypted storage
Inter-agent communication Peer-to-peer protocol that supports identity, payments and dispute resolution