alan-turing-institute/data-safe-haven

Perform data science on data that remains in someone else's server

The main vantage6 repository: code for the central server, nodes, CLI and Python Client

A collaborative learning framework for empowering biomedical research

An Open Framework for Federated Learning.

Noisegate: a differential privacy gateway that lets an untrusted LLM agent query sensitive data over MCP (Model Context Protocol), with a formal guarantee no individual's record can leak even if the agent is adversarial - enforcement lives in trusted code below the model, validated by a runnable attack gallery.

Concrete ML: Privacy Preserving ML framework using Fully Homomorphic Encryption (FHE), built on top of Concrete, with bindings to traditional ML frameworks.

Generates synthetic data and user interfaces for privacy-preserving data sharing and analysis.

Google's differential privacy libraries.

Opinionated setup for securely using AI Platform Notebooks.