Skip to main content

AIA Chain Documentation

Welcome to the technical documentation for AIA Chain - a decentralized AI inference network built on the SENEX Intelligent Chain.

Early MVP experiment — distinct from the shipped AIA product

AIA Chain is an early proof-of-concept for permissionless, browser-native AI inference. The shipped AIA product (the local-first agent and the intelligence-economy loop documented elsewhere on this site, and downloadable from senexic.io) has since evolved well beyond this experiment. These pages are kept for reference and may not reflect the current product.

What is AIA Chain?

AIA Chain is a permissionless, blockchain-based AI inference network that enables:

  • Secure, encrypted AI inference for contributors
  • Transparent, permissionless validation with economic incentives
  • Privacy-preserving communication using end-to-end encryption
  • Browser-native AI models powered by WebLLM

Quick Start

Choose your path:

Validator, installation, and configuration guides are in progress.

Key Features

🔐 End-to-End Encryption

All contributor messages are encrypted using RSA-OAEP before being submitted to the blockchain. Only the targeted validator can decrypt and process the request.

⚡ Permissionless Validation

Anyone can become a validator. No registration, no staking required. Payment is based on:

  • Speed - Faster compute time = higher reward
  • Quality - Higher accuracy score = higher reward
  • Consensus - Agreement with peer validators = bonus

🌐 Browser-Native AI

Validators run AI models directly in the browser using WebLLM, ensuring:

  • No server dependencies
  • Complete decentralization
  • Transparent execution

💰 Fair Economic Model

Contributors pay a small fee per request, which is distributed to validators based on performance metrics.

Architecture Overview

graph LR
A[Contributor] -->|Encrypted Request| B[Blockchain]
B -->|Event| C[Validator]
C -->|AI Inference| D[WebLLM]
D -->|Response| C
C -->|Encrypted Response| B
B -->|Poll| A

Getting Started

Installation, configuration, and first-request guides are in progress. Start with the Contributor Guide.

Technical Deep Dives

Community