Redacta: Free Online AI-Powered PDF Redaction and Pseudonymisation Tool

Redacta is a free, open-source, online tool that uses AI to automatically redact and pseudonymise (de-identify) sensitive information in PDF documents. Describe what to remove in plain language, and AI automatically detects and redacts matching personally identifiable information (PII) across text and images. No signup, no installation, no watermarks. All processing happens locally in your browser using your own API key. Your PDFs are never uploaded to any server.

How It Works

Upload a PDF document.
Describe what to redact in plain language (e.g. "remove all personal names, email addresses, and phone numbers").
AI (Gemini or OpenAI, using your own API key) analyses the document and identifies matching content.
Choose to permanently redact (black out) or pseudonymise (replace with realistic fictional data).
Download the redacted PDF with original formatting preserved.

Key Features

AI-Powered PII Detection and Redaction: Automatically identify and permanently redact personally identifiable information (names, email addresses, phone numbers, SSNs, addresses, account numbers) using natural language descriptions. No manual highlighting required.
AI-Powered Image Redaction: Detect and redact sensitive images embedded in PDFs, including photos, signatures, logos, and screenshots. Redaction is permanent and irreversible.
Document Pseudonymisation (De-identification): Replace real names, addresses, and identifiers with realistic fictional alternatives instead of blacking them out. Ideal for legal review, compliance workflows, training data, and blind recruitment.
Client-Side PDF Processing: Your PDFs never leave your browser. PDF parsing and redaction use WebAssembly (MuPDF) entirely in-browser.
Bring Your Own API Key (BYOK): Use your own Google Gemini or OpenAI API key. No account creation, no signup, no subscriptions, no watermarks. With BYOK, your data goes directly to the AI provider.
PDF Structure Preservation: Redacted documents maintain their original formatting, layout, fonts, and page structure. True source-level redaction, not visual masking.
Free and Open Source: Redacta is completely free with no usage limits. The only cost is AI API usage from your own key, typically fractions of a cent per document.

Use Cases

Legal Document Redaction

Law firms and legal professionals use Redacta to redact privileged information, client names, case numbers, and confidential details from court filings, contracts, depositions, and discovery documents before sharing with opposing counsel or filing publicly. With BYOK, sensitive legal documents are never transmitted to third-party servers.

CV and Resume Redaction for Blind Recruitment

HR departments and recruitment agencies use Redacta to anonymise CVs and resumes by removing or pseudonymising candidate names, photos, addresses, dates of birth, university names, and other identifying information. This enables blind hiring practices and helps organisations reduce unconscious bias in their recruitment processes.

Privacy and Compliance Workflows

Organisations use Redacta to pseudonymise and de-identify personal data in documents for regulatory compliance workflows. Common use cases include fulfilling data subject access requests, preparing documents for third-party sharing, and anonymising training data.

Healthcare and Medical Record De-identification

Healthcare providers and researchers use Redacta to de-identify and redact protected health information (PHI) and patient identifiers from medical records, clinical trial documents, research papers, and insurance forms.

Financial Document Redaction

Financial institutions use Redacta to redact account numbers, transaction details, and personal financial information from statements, audit reports, and regulatory filings before sharing with external parties.

Education and Research

Universities and research institutions use Redacta to anonymise participant data in research papers, student records, and survey responses for ethical review boards and publication.

Frequently Asked Questions

What is Redacta?

Redacta is a free, open-source, AI-powered PDF redaction and pseudonymisation (de-identification) tool that runs entirely in your browser. You describe what to remove in plain language, and AI (Gemini or OpenAI) automatically detects and permanently redacts matching PII in text and images. No signup, no installation, no watermarks. Your PDFs never leave your device.

How does AI PDF redaction work?

Upload a PDF and describe what you want redacted (e.g. "remove all personal names and addresses"). Redacta uses AI to analyse the document, identify matching content across text and images, and permanently redact or pseudonymise it. The redacted PDF preserves its original formatting and layout.

Is my data safe with Redacta?

Your PDF files are never uploaded to any server. All PDF processing happens client-side via WebAssembly (MuPDF). With your own API key (BYOK), AI requests go directly from your browser to the provider. The free trial tier routes extracted text through a backend proxy.

What is document pseudonymisation?

Pseudonymisation (also called de-identification) replaces personally identifiable information with realistic but fictional alternatives. For example, real names become fictional names, real addresses become fictional addresses. This preserves document readability while protecting privacy. It is commonly used for legal review, training data preparation, compliance workflows, and blind recruitment.

What does "bring your own API key" (BYOK) mean?

With BYOK, you provide your own API key from Google Gemini or OpenAI. Your data goes directly to the AI provider with no middleman, no account to create on Redacta, and you only pay the AI provider's standard API rates, typically fractions of a cent per document.

Can Redacta redact images in PDFs?

Yes. Redacta detects and redacts sensitive images embedded in PDF documents, such as photos, signatures, logos, and screenshots. You can fill redacted image areas with a solid colour or add a descriptive label.

Can I use Redacta for legal document redaction?

Yes. Law firms and legal professionals use Redacta to redact privileged information, client names, case details, and confidential data from court filings, contracts, and discovery documents. With BYOK, sensitive legal documents are never transmitted to third-party servers.

Can I use Redacta for CV or resume redaction?

Yes. HR teams and recruiters use Redacta to anonymise CVs and resumes by removing or pseudonymising names, photos, addresses, dates of birth, and other identifying information. This enables blind hiring practices and helps reduce unconscious bias in recruitment processes.

Is Redacta free?

Yes. Redacta is completely free and open source. The only cost is the AI API usage from your own Gemini or OpenAI API key, which is typically fractions of a cent per document.

How is Redacta different from Adobe Acrobat redaction?

Unlike Adobe Acrobat which requires manual selection of content to redact, Redacta uses AI to automatically detect PII based on your natural language description. It also offers pseudonymisation (replacing data with realistic fakes), runs entirely in your browser with no software installation, requires no signup or subscription, produces no watermarks, is completely free and open source, and supports bring-your-own-key (BYOK) for full data control.