Overview
Overview
Features & Roadmap
Getting Started
Resources

Sentinel

PROOF OF CONCEPT

Sentinel is a "Guardrails as a Service" platform for safe WOG AI deployment, using LLM filters to detect and manage unsafe content. It combines open-source and custom guardrails to address public sector-specific risks.

Key Benefits

Acts as the first line of defence, protecting government AI applications from fundamental generative AI risks.

Safeguards citizen data, preventing leaks and ensuring privacy in AI interactions.

Reinforces public trust, supporting regulatory compliance in government AI use.

Ensures safe public engagement, filtering harmful or inappropriate AI-generated content in citizen-facing channels.

Enables secure, scalable adoption of AI across agencies through a compliant cloud-based platform.

Last updated 22 May 2025

Was this article useful?

A Set of LLM Guardrails That Ensure Safe WOG AI Deployment