Privacy Policy

This policy describes what data PDFint8 processes and why.

1. Controller

The controller is Eugen Ionescu, operating under the trading name PDFint8 as an individual, at Danziger Straße 81, 54516 Wittlich, Germany. Privacy contact: support@pdfint8.com.

2. Data we process

  • Document content. Text extracted from the file you upload, the photographs you take, or the text you paste, together with technical details such as file name, size, page count, detected language and detected document type.
  • Classification data. Before any price or credit is confirmed, the document text is classified to determine which CHECK applies. We store the detected document type, a confidence value, the number of documents detected, and short evidence labels such as the document's own title, up to six headings and up to five structural signals (for example “signature block present”). Evidence labels are short descriptors, not copies of the document's content.
  • Analysis results. The summaries, findings and question answers generated for your document.
  • Payment data. Transaction references and status from our payment provider. Card details are handled by the provider, not by us.
  • Account data, where you create an account: email address and authentication identifiers.
  • Technical data needed to operate and secure the service, such as request logs and error reports.

3. What we do with uploaded files and photographs

The uploaded file itself is not retained after processing: text is extracted in memory and the original file is discarded. The extracted text and the analysis results are stored so you can return to your result page and ask questions about the document. Your result is reachable only through the unique private link generated for it, and result pages are excluded from search engine indexing.

Photographs you take with the camera are prepared in your browser, sent to the AI provider so the pages can be read, and are not stored by us: only the resulting text is saved. The same applies to image files and to scanned PDFs that have no text layer — their pages are read as images. Up to 15 photographed or scanned pages are processed per submission, in the page order you confirm.

4. Purposes and legal bases

  • Providing the analysis you requested — performance of a contract with you.
  • Processing payments and preventing abuse — contract and legitimate interests.
  • Keeping the service secure and diagnosing faults — legitimate interests.
  • Meeting accounting and legal obligations — legal obligation.

5. Processors and recipients

  • Hosting and database: the managed cloud backend provided to us through Lovable, where extracted text, classification data and results are stored.
  • AI processing: document text — and, where a document has to be read from images, the page images of photographs, image files and scanned PDFs — is sent to the Lovable AI service at ai.gateway.lovable.dev for reading the pages, classifying the document and generating the analysis. No separate optical character recognition provider is used, and we have no separate contract with an individual model vendor. The underlying model provider, its processing region and whether inputs may be used to train models are determined by that service and are not disclosed in our configuration; we therefore make no claim about them. No account, contact or payment data is sent with document content.
  • Payments: Stripe. Checkout is created on our servers and the payment reference carries identifiers only — the CHECK purchased, the document reference and, for credit purchases, your account identifier. No document text, page image or analysis result is sent to Stripe. Stripe sends the receipt to the email address you give at checkout.

We do not sell your data and do not publish uploaded documents. Access on our side is limited to what is necessary to operate the service.

6. International transfers

Where processing takes place outside your region, it is governed by the data-processing terms of the providers we use — Lovable for hosting, database and AI processing, and Stripe for payments — including the Standard Contractual Clauses where those terms apply them. We do not state a specific processing location, because the processing region is not disclosed to us in the service configuration.

7. Retention and deletion

The uploaded file itself and any photographs you take are processed in memory only and are never written to storage. The text extracted from them, the classification data, the analysis results and any questions and answers are stored for up to 30 days from the moment the analysis is created, and are then deleted automatically by a daily cleanup. Each result page shows its deletion date.

You can delete an analysis yourself at any time before that, using the “Delete this analysis” button on the result page or in your account. Deletion is permanent and removes the extracted text, the free and full results, the classification record and all questions and answers; the result link stops working immediately.

Payment records — transaction reference, amount, status and the item purchased — are kept separately for as long as accounting law requires. They contain no document text, page images, analysis results, questions or answers.

8. Your rights

Subject to applicable law you may request access, correction, deletion, restriction, portability, and object to processing based on legitimate interests. You may also lodge complaint with your supervisory authority. Write to support@pdfint8.com.

9. Cookies

See the Cookie Policy for the technologies used and how to change your choice.

Last updated: 22 September 2026.