Your Data Never Leaves Our Cloud

Meta's Llama is an open-source large language model that can be deployed on any infrastructure. Unlike proprietary models from OpenAI or Google, Llama runs entirely within our own Google Cloud Platform environment. This means we have full control over the model, the data it processes, and where that data is stored.

Open-source AI is the foundation of our privacy-first approach. The model weights are public and auditable, the inference happens on our servers, and no prescription data ever leaves our control.

All AI inference runs on Doses AI's own GCP infrastructure. Patient data is processed within UK and EU data centres that meet the highest standards for data sovereignty and protection. At no point does prescription data travel to external servers or third-party APIs for AI processing.

Our infrastructure is designed with isolation in mind. Each prescription analysis runs in a sandboxed environment, and all data is encrypted both at rest and in transit using AES-256 and TLS 1.3.

Many healthcare AI platforms send patient data to OpenAI, Google AI, or Anthropic for processing. We do not. Doses AI's prescription analysis is fully self-contained. Your clinical data never touches a third-party AI provider's servers.

This is not just a policy choice — it is an architectural decision. Our system is built so that third-party AI calls are physically impossible. The analysis pipeline has no outbound connections to external AI APIs.

Doses AI is designed to meet NHS Digital data handling requirements, including the Data Security and Protection Toolkit (DSPT). Our data processing agreements, retention policies, and access controls align with NHS standards for clinical systems.

We also follow GDPR requirements for the processing of special category health data, ensuring patients have full transparency and control over how their prescription information is used.