Privacy Policy

1. Who processes your data

This Policy describes how the ScanLex project (“we”, the “Operator”) processes data belonging to users of the ScanLex app for Windows and to visitors of scanlex.ru, including the online recognition service at scanlex.ru/online. For any question about data processing you can write to info@scanlex.ru.

By using the app, the website or the online service you agree to this Policy. Personal data is processed in accordance with Russian Federal Law No. 152-FZ “On Personal Data”.

2. Desktop app: your documents stay with you

Not collected

The ScanLex app for Windows performs offline recognition. All text and table recognition happens locally on your device. We do not receive, store or transmit:

• the images, photos, scans and PDF files you open;
• the recognized text or its contents;
• file names or their contents;
• exports to Word, Excel, plain text or PDF.

This data never leaves your computer and is not available to us in any form.

3. Online service: what we receive and how we store it

Sent to the server

For the online service at scanlex.ru/online to recognize your document, that document has to be uploaded to our server. We try to keep this as transparent as possible:

• Transfer is protected by TLS (HTTPS) encryption — the file cannot be intercepted in transit.
• Recognition happens on a server in the Russian Federation. Document recognition itself is performed on our server in Russia. Exception: the optional AI text-improvement feature (see below) — when you start it, the recognized text is sent to OpenAI (USA), i.e. it crosses a national border. This does not affect document recognition itself.
• Recognition is fully automated. No human reviews the contents of your files; they are not used to train machine-learning models and are not used for advertising or profiling.
• AI text improvement — your choice, and text only. After recognition you may voluntarily click “Improve text with AI”. The recognized text (without the source file or image) is then sent to OpenAI (a company in the USA) to correct recognition errors and formatting, and is returned to you. Under OpenAI’s API terms, data sent this way is not used to train its models. The feature is optional: if you do not start it, no text is sent to OpenAI. The Windows desktop app does not have this feature at all — it is entirely offline.
• No registration required. We do not ask for your name, phone number, email address or any other personal data.
• Files are deleted automatically. The source file and the recognition results are kept in a temporary job folder for up to 7 days, so that you can download the finished result again. A copy of the source file is additionally kept in an internal archive for up to 60 days — this is needed to investigate complaints about recognition quality and to fix engine defects. After those periods the files are irreversibly deleted by a daily scheduled job.
• Technical logs (upload time, file size, file name, IP address, selected output formats) are kept for a limited period and used only to diagnose errors and to protect the service from abuse (for example, automated bots).

If you upload documents containing other people’s personal data (for example, copies of passports or contracts), you as the user of the service are responsible for having a lawful basis to process them.

4. Licensing system data (desktop)

Sent to the server

To operate the licensing system of the ScanLex app for Windows (trial period and activation), the app contacts our server and sends the minimum necessary technical data:

• A technical hardware fingerprint — an irreversible hash (SHA-256) of your computer’s characteristics. It is needed to tie the licence or trial period to a device and to prevent copying. This hash cannot be used to reconstruct your PC configuration and cannot identify you as a person.
• App version and operating-system version — for compatibility, diagnostics and correct delivery of updates.
• Computer name — a technical marker for activation logs.
• Anonymised launch statistics (the fact of a launch, licence state) — so we can judge how stable the product is.

The app does not require registration, does not ask for a name, phone number or address, and does not create user accounts.

5. What the website collects

When you visit scanlex.ru, the following may be processed:

• Web-analytics data — we use Yandex.Metrica, which uses cookies to collect anonymised information about the visit (IP address, device and browser type, referral source, actions on pages). This is used to assess how usable the site is.
• Support enquiry data — if you email us, we receive your address and the text of your message and use them only to reply.
• Purchase intent — when you click a purchase button, an anonymised note about the chosen plan is recorded (with no personal data).

You can disable cookies in your browser settings or use the Yandex.Metrica opt-out tool.

6. Why we use the data

• to provide the online recognition service (accepting the file, processing it, returning the result);
• at your request — to improve recognized text with AI (sending the text to OpenAI);
• to provide the trial period and activate desktop-app licences;
• to protect against unlicensed use and automated abuse of the service;
• for technical support and answering enquiries;
• to diagnose errors and improve the stability of the app and the service;
• to understand site usability and product quality.

We do not use the data for advertising or profiling and do not sell it. The contents of documents uploaded to the online service are not used to train machine-learning models.

7. Sharing with third parties

We do not share data with third parties, except for:

• Yandex LLC — for anonymised web analytics (Yandex.Metrica) on the website;
• OpenAI (USA) — only if you yourself start the optional AI text-improvement feature in the online service: the recognized text is then sent to OpenAI for processing (see section 3). If you do not start that feature, no text is sent to OpenAI;
• cases expressly provided for by the legislation of the Russian Federation.

The contents of your documents are never shared with third parties when you use the desktop app (we never receive them in the first place) or during ordinary recognition in the online service. The only exception is the voluntary AI text-improvement feature, which sends the recognized text to OpenAI (see above).

8. Storage and security

• Online-service files are stored on our server in a temporary job folder for up to 7 days; a copy of the source file is kept in an internal archive for up to 60 days. After these periods they are deleted automatically by a daily scheduled job.
• Technical licensing data (see section 4) is kept for as long as the licence needs it and is deleted once it is no longer required.
• Local licence files are stored on your computer in your user profile folder and are not transmitted to us.
• Data transfer between the app or the browser and the server is protected with TLS (HTTPS) encryption.
• Access to the server is restricted by administrative controls and SSH-key authentication.

9. Your rights

Under 152-FZ you have the right to:

• request information about the processing of your data;
• demand that data be corrected, blocked or deleted;
• withdraw your consent to processing;
• request early deletion of a file uploaded to the online service, before the automatic clean-up period expires.

To exercise these rights, write to info@scanlex.ru. We will reply within the period established by law.

10. Children

The product is not specifically intended for children and does not deliberately collect data about persons under 18.

11. Changes to this Policy

We may update this Policy from time to time. The current version is always available at this address, and the date of the last update is shown at the top of the document.

12. Contact