Privacy gateway for AI

The prompt leaves.
Your secrets don't.

OwnKeep sits between your team and ChatGPT, Claude or Gemini. It masks names, contact details and identifiers before a prompt ever leaves the machine, then restores them in the reply. Detection runs locally - no sensitive data is sent away to be checked.

Built for Australian legal and accounting firms. On-premise, local-only masking.

How it works

Three steps, all on your own infrastructure. Staff keep using the AI tools they already know.

1

Intercept

OwnKeep quietly sits in front of the AI service. Every prompt your staff send passes through it first.

2

Mask on-device

Personal information is detected and replaced with placeholders by a model running on your own hardware. The real values never leave.

3

Restore

The AI answers using the placeholders. OwnKeep swaps the real values back in, so your team sees a normal, complete reply.

What you get

A safety layer designed for regulated work, not a checkbox.

Detection stays on your machine

The model that finds personal data runs locally. No prompt, and no PII, is sent to a third party to be scanned.

Australian identifiers

Names, emails, phone numbers, addresses, and TFN, ABN and Medicare numbers - validated, not just guessed.

Fail-closed by default

If the shield can't check a prompt, the prompt is held back rather than sent unprotected. Safety is the default, not the exception.

Compliance-ready audit trail

A record of what was intercepted - the type and count of items - without ever storing the raw values in the log.

Blocks risky file uploads

Attaching a document would bypass text masking, so OwnKeep stops those uploads before the file can reach the AI provider.

Works with the big three

ChatGPT, Claude and Gemini, through the same gateway. Your team keeps the tools they already use.

Made for regulated firms

Legal and accounting practices handle exactly the data that must never end up in someone else's model. OwnKeep lets your people use AI without putting client confidentiality at risk.

  • Client names, matters and file notes stay private
  • Tax File Numbers, ABNs and Medicare numbers are recognised and masked
  • Runs on-premise - your data never leaves your control
  • A clear audit trail you can show a regulator
"From the AI provider's side, it only ever sees placeholders. The real client details never leave the building."
The OwnKeep promise

See it on your own data

Book a short pilot and watch a real prompt get masked, sent and restored - live, on your machine.

Request a pilot