Security

Keep your data yours — and answer safely

Put AI on your public site without the brand, privacy or compliance risk. Concierge answers only from your content, runs on your choice of model, and gives you access controls and a full conversation audit trail.

Trust by design

Security teams ask three questions. Concierge answers all three.

Where do the answers come from? Which AI model touches our data? Who can see the conversations? Concierge is built so you can answer each one with confidence.

Why a buyer cares: an AI on your public site is a brand, privacy and compliance surface. Concierge's grounding, model choice and audit trail make it defensible — to customers, leadership and your security reviewers.

Security at a glance

Your booking data and customer questions stay yours — never used to train models

Grounded only in your knowledge base No training on your data Strict remit — no invented answers Your choice of LLM provider Encrypted in transit & at rest Role-based access controls Full conversation audit trail Token & cost accounting per workspace
How Concierge protects you

Five controls that keep AI support safe

Your content only

Concierge answers strictly from the knowledge you provide. It won't pull in random internet content or invent policy — reducing the risk of off-message or non-compliant replies.

Choose your model

Run Concierge on the LLM provider your business has approved — OpenAI, Anthropic, Gemini and others — and switch without re-platforming if your policy changes.

Access control

Manage who can edit knowledge, change branding and read conversations with role-based access to the dashboard.

Conversation audit

Every conversation is stored with its history, so you can review what Concierge said, spot gaps in your content and demonstrate oversight.

Usage & cost visibility

Token and cost accounting per workspace gives finance and ops clear visibility into AI usage — no surprise bills.

Graceful deflection

Out-of-scope or sensitive questions are declined politely and can route to a human — Concierge never bluffs an answer it can't ground.

Bring your security questions

Book a walkthrough and we'll talk through data handling, model choice and access control for your deployment.