OCR privacy and security
clear data handling for OCR uploads, API calls, saved history, deletion, and enterprise review workflows.
Designed for sensitive documents, invoices, forms, IDs, statements, and internal PDFs.
How OCR data is handled
The product separates uploaded source files, normalized OCR results, account history, API logs, and deletion controls.
Source file processing
Files and image URLs are validated for type, size, and page limits before recognition. The OCR service processes file content only to generate text, Markdown, page JSON, and layout regions.
Saved OCR history
Signed-in web OCR results are saved to OCR History so users can revisit, download, retry, or delete them. Guest submissions do not create account history.
User deletion
Users can delete saved OCR history from Activity > OCR History. Deletion removes the record from normal history views and prevents future user access to that result.
Admin deletion
Administrators with OCR permissions can delete OCR task records from the operations dashboard when handling privacy requests or abuse review.
API data boundaries
API calls store request metadata such as status, latency, credits, source type, IP, and user agent for metering and abuse monitoring. API responses are returned to the caller and charged only after successful OCR.
Operational safeguards
Limits for file size, page count, daily usage, request interval, async cancellation, retries, and upstream circuit breaking reduce overload and repeated provider failures.
Privacy, deletion, and enterprise FAQ
Practical answers for users and teams uploading sensitive documents.
Use OCR with clear data controls
Start with the OCR workspace, review saved history, or integrate with the API while keeping deletion and monitoring paths visible.