Your material is yours. Plainly.
A research library only works if you can put your whole working life into it without hesitation. This page states, in plain words, how Thought Process treats what you keep here. The binding version lives in our terms and privacy policy; this is the approach behind it.
Never training data
Your uploaded sources, notes, and writing are yours alone and are never used to train models. Third-party AI runs only when you connect it, under your own agreement with that provider.
Yours to take
Your library exports in full — sources, annotations, links, and writing. No lock-in, nothing held hostage.
Invite-only visibility
Nothing you keep is visible to anyone you haven't invited. Collaborators see the collections you share with them; everything else stays private to you.
Your model, your plan
AI integration goes through our MCP server: the model is yours, so its data handling follows your own plan with that provider — not a bundled arrangement you never chose.
Run it where your material has to stay.
For teams working with confidential source material, the same workspace runs as a dedicated, isolated instance — and because the library is exposed over MCP, your own AI works across it without the material ever leaving your boundary.
- Dedicated, isolated instance for sensitive corpora
- Your corpus never mingles with anyone else's
- MCP server: your own agents and pipelines, over your own data, inside your boundary
- White-glove setup — we help you stand it up and bring the team in
Questions we haven't answered?
Write to info@thoughtprocess.us — a person reads it.