Privacy
Short version: your journal never leaves your phone, there is no account, and we cannot read anything you write.
Last updated 3 September 2026.
What we store about you
Nothing. Med-Mind has no accounts, no sign-in, and no user database. We do not know your name, your email, or that you exist.
Your journal and your conversations
Everything you write is stored on your device and protected by iOS Data Protection, the encryption built into the operating system. It is never uploaded, never backed up to us, and never used to train anything. It is also excluded from iCloud and iTunes backups, so it does not leave the phone that way either. If you delete the app, that data is gone — we hold no copy that could be recovered, because we never held one.
One thing worth being straight about: iOS Data Protection is keyed to your device passcode. If you have no passcode set, that protection is much weaker. Setting one is the single biggest thing you can do for the privacy of anything on your phone, this journal included.
Which entries you open, how often you write, and how long you spend are all local to the phone and are never transmitted.
The two modes, and what each one sends
| Mode | What leaves your phone |
|---|---|
| On this phone | Nothing. Replies are generated on the device. The app works in flight mode. The only network request it ever makes is the one-time model download described below. |
| Online | The message you send, and enough of the recent conversation for the reply to make sense, go to our server, and from there to Anthropic, whose Claude model writes the reply. They are used for that one request. We store none of it, and Anthropic does not use API traffic to train its models. Your journal as a whole is still never uploaded, and nothing identifying you is sent with it: no name, no email, no device id. |
This changed in version 1.3 (October 2026). Online replies used to be written by our own model on our own machine. They are now written by Anthropic's Claude, because it is a markedly better companion. On-device mode is unaffected and still sends nothing anywhere, and it remains free. If you would rather no company but us ever saw a line of it, use on-device mode — that is exactly what it is for.
You choose between these the first time you open the app, and you can change it in Settings at any time. Online mode is never selected for you.
The model download
If your phone needs the on-device model, it is downloaded once from
medmind.cstsolution.com. Like any web request, our server sees
your IP address and the time in its access log. It does not see, and cannot
infer, anything about what you write.
Analytics, ads and third parties
No analytics SDK, no advertising, no crash reporter that reads your content, no social login.
There is exactly one third party, and only in Online mode: Anthropic, which generates the reply. It receives the message being answered and the recent turns of that conversation, and nothing else — not your journal, not an account, not an identifier. Its API terms say customer traffic is not used for training. In on-device mode there is no third party at all, because there is no network request.
Children
Med-Mind is not directed at children and we do not knowingly collect anything from anyone, of any age.
Your rights
Because we hold no personal data, there is nothing for us to export, correct or delete on request. Deleting the app removes everything. If you would like to ask about any of this, message WhatsApp +84 904 879 900.
Changes
If this policy changes in a way that affects what leaves your phone, the app will tell you in the app before the change takes effect — not only here.