Dreamalickish

Dreamalickish Privacy Policy

Effective: September 15, 2026 Who we are: CtrlAltBreak LLC ("we", "us"). Contact: [email protected].

Dreamalickish is a dream journal that runs its AI on your phone. This policy describes what the app does with information, which is mostly: keep it on your phone.

We also publish a separate Consumer Health Data Privacy Policy covering the one kind of information that can leave your phone — the dream tags you may choose to share. Washington's My Health My Data Act asks for that document to stand on its own, so it does.

The short version

• There is no account. We do not know who you are.
• Your dreams, the readings, and your voice recordings stay on your phone. They are never sent to us or to anyone else.
• The only thing that can ever leave your phone is a short list of abstract tags about a dream (for example "water", "being chased") plus the date — and only if you separately choose to share it, after seeing exactly what will be sent.
• Shared tags are not linked to you, your device, or your IP address. We cannot find them again afterwards, so we cannot delete them on request. We say this before you share.
• We do not sell information about you, and we never hand the tags you send to anyone. We may license aggregate, de-identified trend statistics — how many people, across everyone, dreamt about "water" on a given day — to others through a data API. Nothing in those statistics is about any person.
• We do not show ads. We use no analytics or tracking.
• Dreamalickish is for people 13 and older.

1. What stays on your phone

Everything you create in the app is stored only in the app's local database on your device:

• the dreams you type or speak,
• the readings (interpretations) the app generates,
• the tags the app generates if you choose to share a dream,
• your settings, including your chosen interpreter, reminder time, and your consent choices (with the version of the wording you agreed to and when).

The AI models that produce readings and tags run entirely on your phone. No dream text is sent to any server to be processed. Voice recordings are transcribed on your phone and the audio is not stored or transmitted.

Deleting the app deletes all of this. We have no copy.

2. What can leave your phone, and when

2.1 One-time model downloads

On first run the app downloads two AI models: a language model from Hugging Face (huggingface.co) and a speech-recognition model from GitHub (github.com). Those services see an ordinary download request from your phone — your IP address and basic device information, as any website does — under their own privacy policies. We receive nothing from these downloads. After they complete, the app needs no internet connection to work.

2.2 Dream tags you choose to share (The Collective)

If, and only if, you choose to share a dream, the app sends us this and nothing else:

{ "schema": 1, "date": "2026-08-17", "tags": ["water", "flying"], "checksum": "…" }

• date is the calendar day of the dream, in your local date. No time of day.
• tags are short abstract words or phrases generated on your phone.
• checksum lets the server detect a corrupted upload.

The app shows you this exact payload before every share, and nothing is sent until you approve it. You are asked for your consent to share once, separately from these terms, the first time you share — see the Consumer Health Data Privacy Policy for the full description, your rights, and how to withdraw.

What is never sent: your dream text, the reading, audio, names, places, time of day, location, any account or device identifier, your IP address as part of the contribution, or anything else.

Why: shared tags are combined with everyone else's to show aggregate trends — what people are dreaming about — in the part of the app called The Collective, and in the aggregate trend statistics we make available outside the app (section 2.5). Nothing else.

2.3 Problem and crash reports (optional, previewed)

If the app hits an error, or if you choose "Report a problem" in Settings, you can send us a report. It contains: the error message and the stack trace (only lines naming a place in the app's code — anything else is removed before it is shown to you), which screen you were on, your phone's model and operating-system version, the app version, and any note you choose to write. The app shows you the exact report before it is sent, and nothing is sent unless you tap send. Reports never include your dreams, readings or voice, and nothing in them identifies you; the server does not store your IP address with a report.

2.4 Viewing The Collective

Sharing a dream earns a temporary access token, stored on your phone, that lets the app read the aggregate trend views for about a day. The token is a signed, random string. It contains no information about you and no information about which dream you shared, and our server keeps no record of which token was issued for which contribution.

2.5 Aggregate trend statistics outside the app (the data API)

The same kind of aggregate trend that The Collective shows inside the app — how many contributions on a given day carried a given tag, which tags are rising or fading — may also be made available to others (for example journalists, researchers and developers) through a paid data API. This is how the app is funded: the app itself is free, with no ads and no in-app purchases.

What the API provides is statistics, never contributions: counts and trends across everyone, and only for days on which enough separate people have contributed that no single contribution can be picked out. Individual contributions are never returned by the API, and as section 3 describes, there is nothing in a contribution that could identify a person in any case.

We treat these statistics as de-identified data in the sense of Washington's My Health My Data Act, and we commit publicly to the three things that definition asks for:

• we take reasonable measures to ensure the statistics cannot be associated with any person — no identifier exists to begin with, dates are day-level only, and no figure is published for a day until a minimum number of separate people have contributed;
• we process and publish them only in de-identified form and will never attempt to re-identify anyone; and
• everyone who receives them through the API is bound by contract to the same: not to attempt re-identification, and to pass that obligation on to anyone they share the statistics with.

3. IP addresses and server logs

When your phone talks to our server (to share tags or to read The Collective), the server necessarily sees your IP address in order to reply. We use it only to deliver the response and to protect the service from abuse (for example, rate limiting). It is not stored with any contribution and there is no place in our database to store it. Ordinary short-lived infrastructure logs may contain it, as with any internet service; they are not joined to contributions and are not used to identify anyone.

4. What we do not do

• No accounts, no sign-in, no email collection.
• No sale of information about you, and no handing over of contributed tags to anyone. The only thing that ever leaves our server is aggregate statistics (section 2.5), shown inside the app or provided through the data API.
• No advertising, no ad SDKs, no ad identifiers.
• No analytics or tracking SDKs, and no third-party crash-reporting service. Crash reports exist (section 2.3) but go only to us, only when you send them, and only after you have seen them.
• No location collection, and no geofencing of any kind.
• No cloud AI. No dream is ever processed by a remote model.

5. Children

Dreamalickish is not directed at children under 13, and we do not knowingly collect information from them. The app asks for a birth year on first run and does not admit anyone under 13; the year itself is not stored. If you believe a child under 13 has shared tags through the app, contact us at [email protected]. Because contributions are not linked to a person, we are unable to identify or remove a specific child's tags, but we will respond and act on any information we can.

6. Your rights and choices

• See what would be shared: every dream's exact payload is shown in the app before it is sent, and afterwards in the journal.
• Decline or withdraw sharing: sharing is off by default, per dream. You can withdraw your sharing consent at any time in Settings, which stops all future sharing. The journal and readings are unaffected.
• Access and deletion of what is on your phone: it is on your phone. Deleting the app deletes it.
• Access and deletion of shared tags: contributed tags are not linked to you, by design, so there is no way for us to locate "your" tags to return or delete them. This is the trade we make for genuine anonymity, and it is why the app says "sharing is permanent" before you agree.

Residents of some US states have additional statutory rights. See the Consumer Health Data Privacy Policy for the Washington and Nevada process, including how to appeal.

7. Where we operate

The app is offered in the United States. Our server is hosted in the United States (DigitalOcean), with Cloudflare in front of it for connection security and abuse protection; Cloudflare sees the same request metadata any network intermediary does and none of it is stored with contributions. We do not offer the app in the European Economic Area or the United Kingdom.

8. Changes

When this policy changes, the effective date above moves. If the change affects what is collected or why, the app will ask you to agree again before anything is shared under the new terms.

9. Contact

CtrlAltBreak LLC [email protected]