Skip to main content

Introduction

PADO leverages the AO Computer on the AR Network to offer zkFHE-based decentralized computing units, providing trustless and confidential computing capabilities for the AO ecosystem. It also utilizes Arweave as a private data storage layer.

Users who want to share their encrypted private data from Arweave can easily utilize PADO's zkFHE techniques for confidential computation requests within the AO ecosystem. These requests are sent from AO to PADO's zkFHE computation nodes.

PADO's zkFHE techniques are designed on top of the FHE algorithm, which synchronously performs ZK proof on the homomorphic computation process. The simultaneous integrity constraints ensure the reliability of homomorphic operations and the correctness of the encrypted result while preserving data privacy during the whole process.

zkFHE is a novel cryptographic technology that organically combines the two cutting-edge cryptographic primitives, ZK and FHE, and uses the advantages of the two to complement each other, thus providing practical verifiable confidential computation(VCC). PADO's zkFHE technique offers several core advantages, including openness, privacy-friendly, and computational integrity. We believe that zkFHE is the fundamental technology for building a verifiable data economy in the Web3 era.

PADO's zkFHE network integrates seamlessly with the AO and Arweave networks. As long as the computing, scheduling, and storage functionalities operate independently, this allows for the indefinite expansion of computing power. This enables the AO ecosystem to leverage PADO's zkFHE technology to access verifiable confidential computation and comprehensive privacy protection capabilities.

About the Implementation​

Note in this particular scenario built on AO, we use a linearly homomorphic encryption (LHE) scheme for data-sharing purposes. The initial version of the prototype provides a plain threshold FHE version to achieve the whole functionalities without zk-based integrity ensurance. However, we believe this basic algorithm meets most of the requirements for simple applications. We will gradually build this repo with zk-LHE, adding a zero-knowledge capability to prove each procedure of the underlying linearly homomorphic encryption scheme.

Main Components​

The data sharing and incentive workflow based on AO allows data providers to share their data securely and privately with data users. Key components are listed as follows:

  1. PADO AO SDK

    Through the SDK developers can upload a user's encrypted data, and decrypt the ciphertext data.

    For specific interface details, please refer to the github link: PADO AO SDK.

  1. PADO AO Process

    PADO AO Process is created on AO, which mainly manages data, nodes, verifiable confidential computing tasks and related results. The process also handles computation costs.

    The github link: PADO AO Process.

  1. PADO Node

    PADO Node is an environment that truly performs verifiable confidential computations. Mainly to obtain verifiable confidential computing tasks, execute tasks, and report results.

    The github link: PADO Node.

    The WASM wrapper for threshold-zk-LHE.

Workflow​

proofs

  1. Register PADO Node

    After the PADO Node is started, it shall be registered in the Node Registry Process in AO. The registered information includes name, description, public key, owner address, etc.

  1. Upload data

    Data Providers can upload encrypted data through dapp based on the PADO SDK and set data prices at the time of upload. The data encrypted by the FHE algorithm and the PADO Node public key will be uploaded to Arweave, and the data information will be registered to AO's Data Registry Process.

  1. Submit task

    Data consumers can submit computation tasks with their public keys through dapp based on PADO SDK, and pay certain computation fees and data usage fees. The computation tasks will be submitted to AO's Task Process.

  1. Task execution

    PADO Node obtains computing tasks from the Task Process in AO, uses the LHE algorithm to compute the tasks, and reports the results to the Task Process after computation. Task Process verifies the results. After completing the verification, the fee is distributed to the data provider and PADO Nodes.

  1. Get Result

    Data consumers obtain encrypted data from Arweave, obtain task results and related information from the Process, and then use the LHE algorithm with their private key in the SDK to decrypt the results.