Privacy

Overview

UnderstandDocs analyzes pasted text and uploaded documents to generate summaries and structured insights.

Supported formats: TXTDOCXPDF

For supported formats, text extraction is performed locally in your browser whenever possible. Raw files are not automatically uploaded as part of the standard extraction flow.

Privacy-First Processing

Before AI analysis, common sensitive values may be automatically detected and masked. This can include identifiers, contact details, and payment or account references.

Masking is applied as a best-effort measure. It is not a guarantee that all sensitive data will be caught.

AI Processing

After extraction and any applicable masking, text may be securely sent to the Anthropic API to generate your results.

No user profile is built from submitted documents. Content is used solely to produce the requested output.

Data Retention

UnderstandDocs does not maintain a saved document library or permanent workspace storage.

Limited technical metadata may be logged for abuse prevention and service reliability. This may include request timestamps, hashed or anonymized identifiers, request size, response times, and error categories. Full submitted document text is not logged.

Your hosting provider or infrastructure services may maintain their own operational logs independently of this application.

Analytics & Session Replay

UnderstandDocs uses privacy-focused analytics tools to improve usability, reliability, and overall user experience.

We may collect anonymous usage information such as page interactions, clicks, scrolling behavior, general browser/device information, and technical diagnostics related to site performance and reliability.

To better understand usability issues and improve the product experience, we may use session replay analytics tools such as Microsoft Clarity.

Sensitive document content is intentionally masked from analytics recordings.

Uploaded document text, extracted text, filenames, document previews, and AI analysis results are not recorded in session replay analytics.

We do not intentionally use analytics tools to monitor or store the actual contents of documents submitted to UnderstandDocs.

Analytics providers may still collect limited technical metadata such as browser type, approximate geographic region, device information, interaction timing, and anonymized usage patterns as part of their platform functionality.

Sensitive Documents

If your document contains highly confidential, legal, financial, medical, or regulated information, exercise caution before submitting it to any AI processing service — including this one.

When in doubt, consider redacting sensitive details manually before analysis.

Accuracy Notice

AI-generated outputs may contain errors, omissions, or incomplete context. Always verify important information independently before making decisions based on results.

Contact

For privacy questions or concerns, contact us at hello@understanddocs.com.