A privacy checklist before using online OCR for sensitive documents

Confirm that you may process the file

Ownership of a copy does not always mean permission to upload it to a third-party service. Workplace policies, client agreements, professional duties, and data-protection rules may impose additional limits.

If the answer is unclear, use an approved internal system or ask the responsible organisation before uploading.

Minimise the document first

Remove pages and fields that are unrelated to the goal. If you only need text from one page, do not upload an entire personnel file. Redact identifiers only with a method that truly removes the underlying content rather than drawing a black rectangle over it.

Use a non-sensitive sample when testing whether a workflow produces the desired output.

• Remove irrelevant pages
• Redact unnecessary personal identifiers
• Avoid uploading passwords or private keys
• Test the workflow with synthetic data first

Check hosting and retention

A responsible service should explain where processing happens, how long uploads and results remain available, and how users can contact the operator. Short retention reduces exposure but does not replace a lawful basis or internal approval.

Temporary result links should be treated as confidential. Download the output, store it in the approved location, and avoid sharing the URL in public channels.

Know when local processing is the right choice

Highly regulated, secret, or legally privileged material may require local OCR or a private deployment with contractual safeguards. An online public tool is not automatically appropriate just because it is technically capable of processing the file.

For ordinary business documents that you are permitted to process, a clearly documented EU-hosted service with automatic deletion can be a practical option.