LectureHub is designed for B2B deployments where the content is sensitive, the audit trail matters, and the AI model can't see anyone else's data.
What the hub does with your data, who can see it, and where it runs. The trust page carries each control with its status; this is the short form.
Your content is isolated by tenant. No training on customer data, ever.
Run on a shared model, a dedicated instance, or your own hosted LLM.
The assistant answers from the sources you've explicitly admitted to the hub.
Granular access by team, role, and content collection. Inherits from your IdP.
Every answer cites the document, recording, or person it came from.
Configurable retention windows for prompts, answers, and uploaded content.
EU or US regions, single-tenant deployments, and bring-your-own-cloud available.
We list certifications only after they're verified. Ask us for the current state.
Access by team, role and content collection, inherited from your identity provider. The rule holds for the assistant too: it never answers from a collection the person asking cannot open.
| Role | Handbook | Support runbooks | Pricing & discounts | Board papers |
|---|---|---|---|---|
| Everyone | ||||
| Support | ||||
| Sales | ||||
| Leadership |
Three ways to run the model behind the assistant. In every one, your content trains nothing — ours or anyone else's — and every answer cites its source.
Our managed model, tenant-isolated. The fastest way to start.
The question and the approved passages needed to answer it. Nothing is kept for training.
A model instance that serves your hub and no one else's.
The same, to an instance that serves only you.
Bring the LLM you already run — in your cloud or ours. The hub talks to it and to nothing else.
Nothing. The model runs where you run it.
Bring your security questionnaire. We answer it in writing, and show you the hub on your own data before you decide.