Back to SafeRag

SafeRag Privacy Policy

🔒 100% Local Processing - Zero Cloud Exposure

Last updated: November 15, 2025

Privacy-First by Design
SafeRag is built with absolute privacy as its foundation. All AI processing happens locally on your Mac using Ollama. No data ever leaves your device. No cloud connections, no tracking, complete control. Your conversations, documents, and data remain entirely yours.

Overview

SafeRag ("we", "our", or "us"), developed by Corixa, operates a macOS application for private AI conversations with RAG (Retrieval-Augmented Generation) capabilities. This privacy policy explains our commitment to your privacy and how SafeRag handles (or rather, doesn't handle) your information.

Our Privacy Commitment

SafeRag was created specifically for privacy-conscious professionals who need AI capabilities without cloud exposure. We adhere to three core principles:

Local-First Architecture: All processing happens on your Mac
Zero Data Collection: We don't collect, store, or transmit your data
Complete Transparency: Open about every aspect of data handling

What SafeRag Does NOT Do

To be absolutely clear, SafeRag does not:

How SafeRag Works

Local AI Processing (Ollama)

SafeRag uses Ollama, an open-source local AI runtime, to provide conversational AI capabilities:

Local Data Storage

All your SafeRag data is stored locally on your Mac using Apple's secure frameworks:

Data location: ~/Library/Application Support/io.halldor.SafeRag/

RAG Document Intelligence

When you upload documents for RAG (Retrieval-Augmented Generation):

Supported Document Types

All processed locally without external services:

Multi-User Accounts

SafeRag supports multiple users with role-based access control:

Watch Folder Automation (Pro Feature)

When using watch folder auto-ingestion:

Compliance Features

GDPR Compliance

SafeRag helps you comply with GDPR through local-first architecture:

HIPAA Compliance

For healthcare professionals, SafeRag supports HIPAA compliance:

Audit Logging (Pro Feature)

For compliance and security monitoring:

Data Security

Your data is protected through multiple security layers:

Network Activity

SafeRag has minimal network requirements:

Ollama Model Downloads (One-Time)

The only network activity occurs when downloading AI models:

In-App Purchase Verification (Optional)

If you purchase SafeRag Pro via the Mac App Store:

No Analytics or Telemetry

Unlike most apps, SafeRag does not:

Your Rights and Control

You have complete control over your SafeRag data:

Data Portability

Data Deletion

Data Retention

Third-Party Services

Ollama (Required)

SafeRag requires Ollama for AI functionality:

Apple StoreKit (Optional)

If purchasing Pro via Mac App Store:

No Other Third Parties

SafeRag does not integrate with:

Children's Privacy

SafeRag is not directed at children under 13. We do not knowingly collect information from children under 13. The app requires macOS 14.0+ and is intended for professional use by adults.

Data Breach Notification

Since SafeRag processes all data locally with no cloud storage:

International Data Transfers

Not applicable. Since all data stays on your Mac, there are no international data transfers. Your data never leaves your device regardless of where you're located.

Changes to This Policy

We may update this privacy policy to reflect changes in SafeRag's functionality or legal requirements. Changes will be posted on this page with an updated revision date. Material changes will be announced via:

Continued use of SafeRag after changes constitutes acceptance of the updated policy.

Open Source Transparency

While SafeRag's source code is currently proprietary, we are committed to:

Questions About Privacy?

We welcome questions about SafeRag's privacy practices. Our architecture is simple: everything stays on your Mac. If you have specific concerns or need clarification:

Contact Information

For privacy-related inquiries, data requests, or general questions:

Company: Corixa
Product: SafeRag for macOS
Jurisdiction: Iceland

Summary: What Makes SafeRag Different

Most AI tools require you to trust a company with your data. SafeRag is different:

Zero Trust Architecture: You don't have to trust us because we never see your data
Verifiable Privacy: All processing happens locally on your Mac
Complete Control: You own your data, your models, your conversations

This isn't just a privacy policy—it's our product architecture. Privacy isn't a feature; it's the foundation.