ZeroEgress | Remove Sensitive Info From Documents

Requests to other hosts: countingModel: loadingThis device: measured at your first scan

Drop your documents here

drag one or many files, PDF, Word, Excel, images

Output PDF is built from scratch, no metadata from the original. Text, comments, and form fields are scanned. Text inside embedded images is not: drop such images as separate files to OCR them. ZeroEgress scans the text extracted from the document. For Word files (.docx) that is the document body: headers, footers, footnotes, comments, and text deleted under tracked changes are not part of the extracted text, so they are not scanned or redacted. Accept or reject tracked changes, and review headers, footers, and comments separately when they may hold sensitive content.

Or paste text

Drop file here, PDF, Word, Excel, PowerPoint, image, or text file

First time? The AI model downloads in ~1 minute. After that, everything runs in your browser.

Redacted document, then a compliance report: entity summary and audit chain

Paste into AI tools

Save clean text for AI tools

Placeholder ↔ original (sensitive)

AES-256-GCM

Advanced options

Time window per segment. The recommendation above picks one for you; 5000 ms is the default. Recorded as T_max.

Keep specific text from redaction (optional, comma-separated):

Custom entity patterns (optional, Name:regex, comma-separated, max 10, 200 chars each):

Output mode: Redact with labels [NAME], [EMAIL] Tokenize with IDs [NAME_1], [EMAIL_1]

Use tokens if the AI needs to distinguish between entities (e.g., multiple clients in one document).

Annotated Text

Scan a document to see results

Audit Chain NOT VERIFIED

No events yet

How it works

Drop your document

PDF, Word, Excel, image, even scanned.

AI finds what's sensitive

You review and choose what to keep.

Download clean version

Use with any AI safely.

Before and after

Before

John Smith, CEO of Acme Corp, met with Dr. Sarah Chen at john@acme.com or 212-555-1234. SSN: 123-45-6789.

After

, CEO of , met with at or . SSN: .

Your client info stays private. The AI gets what it needs.

Tokens, and document types

Tokenize: keep the shape, lose the details

Replace names and emails with reusable tokens like [NAME_1], [EMAIL_1]. Paste the tokenized text into ChatGPT. Get an answer. Reveal the tokens in the response to restore the original details. Your data never leaves your device.

Auto-detects your document type

Medical record, financial statement, legal brief, cybersecurity report: ZeroEgress recognizes the content and recommends the matching security profile, with its plaintext window, before you scan. One click applies it. Healthcare gets one timing profile, financial another, cybersecurity another, and both the recommendation and your choice are recorded.

Nothing to trust

Nothing leaves your browser.

Every scan can export a signed evidence pack, with a verifiable time bound on when the records were created.

Output PDF has zero metadata from the original.

Export as clean text for ChatGPT/Claude: no formatting, just redacted content.

No account. No login. No signup.

No tracking. No cookies. No analytics.

Works with PDFs, Word, Excel, and images.

Once loaded, it runs with the network off. Reloading the page needs the network.

Share documents with AI without sharing sensitive info.

Annotated Text

Reveal AI response

How it works

Drop your document

AI finds what's sensitive

Download clean version

Before and after

Tokens, and document types

Nothing to trust

Share documents with AI without sharing sensitive info.

Try it

Annotated Text

Reveal AI response

How it works

Drop your document

AI finds what's sensitive

Download clean version

Before and after

Tokens, and document types

Nothing to trust