System Architecture

Protocol Overview

Intract Protocol is a secure protocol which enables users to store all their private data securely using same standards. The data is encrypted and controlled only by the user. Users have complete control, being able to share selected data segments with trusted authorities—such as AI agents—without exposing the raw underlying information. This ensures that any computations are performed securely, and only the processed results are shared.

Architecture Overview

The four hoursemen of Intract Protocol

The architecture comprises of 4 layers (or 4 horsemen). These 4 layers form the core of the protocol that allows us to achieve all of the system objectives. Let's explore each of these layers.

Each of these layers handle a specific aspect of data management. (We will dive deeper into each one later)

1. Ingestion Layer: Data Ingestion Engine (DIE)

   • Role: Securely handles acquiring and importing of user data from various sources (e.g., social media, email) after obtaining user permission.

   • Non-Technical Analogy: Think of this layer as the "data pickup" phase—collecting your information securely.

2. Access Layer: Secure Access Management (SAM)

   • Role: Manages permissions and controls access, ensuring only authorised entities can view or use specific data segments.

   • Non-Technical Analogy: Similar to a security guard who ensures that only the right people have access to your private vault.

3. Storage Layer: ENCRYPTED STORAGE SERVICE (ESS)

   • Role: Enables the user to store their data in an encrypted format, ensuring that even if intercepted, the data remains secure.

   • Non-Technical Analogy: Imagine a high-security vault where your valuables are kept under advanced locks.

4. Compute Layer: TRUSTED EXECUTION CLUSTER (TEC)

   • Role: Executes secure computations on the encrypted data using a decentralised TEE network, exposing only the processed results to third parties.

   • Non-Technical Analogy: Like hiring a trusted expert who can work on your secret recipe without ever seeing the full details—only the final dish is served.

A (very) simplified data flow

The following steps outline the journey of user data through our system:

1. User Authorization:

   • The process begins with the user granting permission to ingest data from various platforms.

2. Data Ingestion:

   • The Ingestion Layer manages collection of this data and prepares it for secure storage.

3. Secure Storage:

   • Collected data is encrypted and stored in the Storage Layer.

4. Access Management:

   • When a third party or an application requests data, the Access Layer verifies the permissions and ensures that only authorized data or computed results are shared.

5. Secure Computation:

   • The Compute Layer performs decentralized, secure processing on the data, providing only the results to the requester without exposing raw data.

Key Architectural Features

• Layered Security: Each layer incorporates specific security measures, ensuring robust protection at every stage.

• Decentralized Computation: Utilizing a decentralized Compute Layer reinforces trust and minimizes the risk of data manipulation or leakage.

• User-Centric Design: The architecture is designed with the user in mind, prioritizing privacy, control, and seamless data portability.

• Scalability & Flexibility: Our modular design facilitates the addition of new data sources and adapts to evolving security challenges.