Data Controls

Effective date: February 11, 2026. This page explains exactly where data lives, what requires explicit permission, and what choices you control.

Default behavior (local-first)

What requires explicit permission

Cloud and sync model

The intended architecture is user-controlled storage: you choose whether to stay local-only or connect your own cloud provider. If enabled, sync should use end-to-end encrypted data blobs so providers cannot read workspace content.

Until that feature is live, default behavior remains local-only.

AI data boundary

AI features should be optional and separately consented. If enabled, only the minimum required payload should be sent to the selected AI provider, and only for the session/use case you trigger.

No AI processing should run on your data unless you explicitly turn it on.

Medical and legal boundary

Divergify is a productivity environment, not a medical provider. We do not diagnose, treat, or provide clinical care. The tools can be used by people with or without formal diagnoses.

If we launch regulated healthcare features in the future, we will add the required compliance controls first.

California + HIPAA quick answers

Public web inference risk

Public pages can still be indexed and observed at the network/browser level. That means third parties may infer topic interest from visited URLs even when workspace data stays local.

We reduce this risk by avoiding ad trackers and keeping workspace content local by default.

Your controls right now

State-law alignment

Divergify is built to support consent-first handling and local-first minimization across US state privacy frameworks. Where laws differ by state, we aim to apply the stricter operational standard and update controls as requirements evolve.

For legal details, see Privacy Policy and Terms of Service.