Legal · Privacy
Privacy.
This page states how Kennan handles your data. The commitments below are binding as written. An expanded formal policy is in legal review and will replace this page; it will not weaken anything stated here.
Effective 2026-06-27 · updated 2026-10-22 · interim policy
What we handle
The documents and data you upload, the files Kennan pulls from a Box or Dropbox account you connect, the read-only access token for that account (held in Google Secret Manager, never in our database), the analyses and deliverables Kennan produces from all of it, and the account, billing, and usage metadata needed to run the service. Your work product is yours — see the Terms.
Where it runs
Kennan runs on Google Cloud Platform, with per-organization isolation across physical storage, database row-level security, and a scoped query layer. The architecture is described on the Security page.
Model and parser vendors
To do the work, Kennan sends content to large-language-model and document-parsing vendors. No vendor in the data path trains on your data. How long each vendor retains the content it processes is governed by that vendor's own terms. The current vendor list — who sees what, under which contract — lives on the Security page.
What we don't do
We don't train models on your data, and we don't sell it.
Your controls
You can export your workspaces at any time and delete them when you're done. If you leave Kennan, the work leaves with you.
Questions
For data-handling questions, email security@kennan.ai or get in touch.