Architecture Overview

The LatticaAI platform enables the secure processing of AI models using homomorphic encryption. The system consists of clients for management and query operations and a backend for interaction management and computation.

Diagram of the main components in the LatticaAI system.

Components

1. LatticaAI Clients

⫸ The Management Client enables submitting an AI model provided by the AI Provider. Once submitted, LatticaAI prepares the model to process encrypted data securely. This preparation creates a homomorphic-ready model in the LatticaAI backend, referred to as a model.

After the model is deployed and prepared for homomorphic encrypted data processing, it is associated with an Access Point, enabling secure interactions and management.

AI Providers can then use the Management Client or Web Console to manage Model access permissions and allocate computing resources for processing user requests

⫸ The Query Client facilitates secure interactions with deployed models.

► It encrypts input data before sending it to the backend for computation.

► The backend processes queries using the homomorphic-ready model and returns encrypted results.

► The client decrypts the results locally, ensuring data privacy throughout the process.

Management Client

Enables AI Providers to control model deployment, manage user permissions, and oversee system resources

Query Client

End users use this client to access the deployed AI models.

---

2. LatticaAI Backend

⫸ Interaction Manager: Routes requests to the appropriate AI model.

⫸ Hardware Acceleration: Currently GPU-based for high-performance processing with adaptability for future hardware acceleration technologies as they become available.

---