Privacy Policy

Last updated: 12 October 2026

BookCat is an e-book reader for iPhone and iPad. This policy explains what happens to your data when you use the app and this website. The short version: your books stay on your devices and in your own iCloud; text goes to an AI provider only when you ask a question, only the part the question needs, and we don't keep it; there are no ads, no tracking and no analytics.

BookCat is made by Lev Litvak, an individual developer ("we", "us"). Contact: support@bookcatapp.com.

What stays on your device and in your iCloud

  • Books, reading positions, bookmarks, your vocabulary and review progress are stored on your device. If you're signed in to iCloud, they are synced through your private iCloud database (Apple CloudKit). That database belongs to your Apple Account; we cannot see what's in it.
  • Reading statistics (pages and words read) are counted on your device and never leave it.
  • Settings and the AI answers you've already received (the answer cache) are stored on your device only.
  • Your own API key, if you enter one, is stored in your device's Keychain and synced through iCloud Keychain. We never receive it.

When you ask the AI

BookCat's smart features — word meaning and translation, explaining a passage, recaps, characters, live chapter translation, tag translation — are answered by OpenAI. Before the first question leaves your device, the app asks your permission. You can withdraw it at any time in Settings → Data and AI.

What is sent: only what the question needs, and only from what you've already read — the word with its sentence, the passage you selected, or the chapters up to your page — together with the instructions for the model.

Which way it goes:

  • Without your own key, the question goes through our server (hosted by Cloudflare), which forwards it to OpenAI and streams the answer back. The server does not store or log the text of your books or of the answers. It asks OpenAI not to store the response (store: false).
  • With your own OpenAI key, the question goes from your device directly to OpenAI. Our server is not involved, and OpenAI's terms for your account apply.

What OpenAI does with it: under OpenAI's API data policy, API requests are not used to train its models. OpenAI may keep requests for up to 30 days to monitor for abuse, then deletes them. See OpenAI's privacy policy.

Nothing you read is ever used to train a model by us.

What our server keeps

To give everyone a free allowance and to count a subscription's monthly limit, our server keeps a few records. None of them contains text of your books, your name or your email.

  • A pseudonymous reader identifier. If you're signed in to iCloud, the app sends a one-time token; the server asks Apple's CloudKit service which iCloud user it belongs to and stores only a keyed hash (HMAC) of that identifier, so that your iPhone and iPad share one allowance. Without iCloud, the identifier is a random name for this installation, stored in your device's Keychain.
  • Device integrity checks. The app proves to the server with Apple's App Attest that it's a genuine copy, and uses Apple's DeviceCheck to mark that a device has already received its free allowance. Apple does not tell us who you are.
  • Usage counters: how many times each feature was used and how many units of the allowance were spent, per day and per feature, linked to the pseudonymous identifier.
  • Operational records: for each answer — the pseudonymous identifier, the feature, the number of tokens, the cost and whether it succeeded — kept for about three months; and a technical log of requests (path, status code, time, a one-word reason for a refusal), kept briefly by Cloudflare.

We use these records only to run the service: to apply allowances and limits, to stop abuse and to understand the cost of the service. We do not sell them, share them for advertising, or combine them with data from other sources.

Purchases

Subscriptions are sold by Apple through the App Store; we never see your payment details. We use RevenueCat to check whether your subscription is active. RevenueCat receives your purchase history, a pseudonymous purchaser identifier (a hash derived from your iCloud account, or an installation identifier) and basic technical information about the device and app version that its SDK sends. See RevenueCat's privacy policy.

Archive of Our Own

If you open or import works from Archive of Our Own (AO3), the app connects to AO3 directly from your device, only when you do so, using the browser inside the app. AO3's own terms and privacy policy apply. BookCat is not affiliated with AO3 or the Organization for Transformative Works.

This website

This website uses no cookies, no analytics and no third-party scripts or fonts. Our hosting provider, Cloudflare, processes technical data such as IP addresses to deliver pages and protect against attacks.

Tracking

BookCat does not track you across apps or websites owned by other companies, shows no ads, and contains no analytics or advertising SDKs.

Legal bases (EEA, UK and Switzerland)

We process data to provide the service you ask for (performance of a contract) — the allowance, the subscription and the AI answers — and for our legitimate interest in preventing abuse and keeping the service running. Sending text to OpenAI happens only with your consent, which you can withdraw in the app at any time.

Your rights

Depending on where you live, you may have the right to access, correct or delete your data, to object to or restrict processing, and to complain to a data protection authority. Because the records on our server are pseudonymous and contain no contact details, we may need your help to find them — write to support@bookcatapp.com and we'll work it out with you.

You can delete everything stored on your device by deleting the app, and the data in your iCloud through your Apple Account settings.

Children

BookCat is not directed at children under 13, and we don't knowingly collect data from them.

International transfers

Our server runs on Cloudflare's global network, and OpenAI and RevenueCat process data in the United States. Where required, these transfers rely on the providers' standard contractual safeguards.

Changes

If this policy changes, the new version will be posted here with a new date. If a change matters for how your text is handled, the app will ask for your permission again.