Skip to content

Security & data flow

Know what goes where.

A privacy claim should have a boundary. Here is what Hushbridge processes locally, what reaches an AI provider, and what still needs your review.

Free privacy check

Browser only

Your text and the swap table remain in this tab. No model call or metering request is made by this preview.

Browser extension

Local swap → existing AI site

Detected values are replaced on send. The AI site receives the protected prompt; credit metering sends a usage count to Hushbridge. The site itself can potentially read text while you type.

Web workspace

Local swap → Hushbridge server → model

The protected prompt passes through the Hushbridge server. The original values and swap table stay in the browser. Account and billing records are separate from prompt content.

SDK / local proxy

Your process → model provider

The engine operates in your process. Your integration controls the data sent to the provider. Metering reports credit usage separately.

Before you send

The context is still context.

Replacing a name does not make the surrounding story anonymous. A rare job title, an incident or a combination of facts can still identify someone.

Hushbridge does not guarantee complete detection. The browser extension does not inspect attachments. Review the protected prompt and follow your organization’s AI policy.

Hushbridge does not offer a HIPAA BAA and does not claim SOC 2 certification. Its output is not a guarantee of legal de-identification.

Read the existing product security documentation ↗

THE LOCAL PROTECTION WORKFLOW

Conceptual diagram

01 / YOUR DEVICE

Find & replace

Client emailConsistent placeholder

The original values stay in your local swap table.

02 / YOUR REVIEW

Check what remains

Instructions + context
+ placeholders

Check missed details and identifying context before sharing.

03 / YOUR MODEL INTEGRATION

Send reviewed text

The provider receives the text you choose to share.

Restore supported replies locally using the same swap table.

The free browser check demonstrates protection and restoration only; it makes no AI request. Sending text and handling replies depend on your chosen workflow. Account and usage data are separate.

Remove sensitive details before your next AI prompt.

Try the free check ↗