Control what is shared.
Find sensitive values and apply the right treatment: tokenise, redact, keep local or block the request.
Kordane BoundSenseInformation control for AI
Your information. Your rules.
Set the terms before information reaches a model. Protect sensitive values, decide where a request can run and keep a record of the decision.
Demonstrated in the browser · available in pilots
Interactive demonstration
Follow one controlled crossing.
Choose a policy, run a request and inspect the result. See what is protected, where the request may go and which decisions are recorded.
User or applicationSensitive contextAI model or toolResponse or action
Policy profile
Sensitive values are tokenised or redacted, codewords may not leave at all, and the request is routed only to an approved external model.
Synthetic demonstration. Runs deterministically in your browser on fictional data. Nothing is transmitted.
Ready. Choose a profile and run the crossing.
Original request
Sanitised outbound request
Policy & audit timeline
Route: the approved external model.
authorisedtransformedblockedcontained
Synthetic model responsewaiting
Restore is a local substitution inside your browser. Reset clears the request, result, and audit trail.
Choose the permitted path.
Define which models, tools and environments are allowed. Require human approval when the action needs it.
Understand each decision.
Review the information found, the policy applied and the destination selected in one trace.
BoundSense Audit
One crossing, fully recorded
The decision record a single synthetic crossing leaves behind.
Operational briefing (synthetic fixture)
- Policy profile
- Controlled Routing
- Detected
- 4 sensitive values, 3 classes
- Transformation
- 3 tokenised · 1 redacted
- Environment
- Private cloud / VPC
- Route
- Approved private model
- Restore
- Tokenised values restored locally
- Audit
- 8 events recorded, reviewable
Every crossing produces a reviewable record: what was found, what policy did, where the request went.
Illustrative decision record · Synthetic data
Follow the policy and information lifecycle
Detect
Sensitive values are found in the request before anything moves: identifiers, codewords, references, contact channels.
Classify
Each value is assigned a class your policy can reason about: identity, document, contact, codeword, reference.
Apply policy
The active profile decides per class: pass, tokenise, redact, block, or contain. Decisions are explicit and versioned.
Transform
Values that may cross in shape but not in substance are tokenised; values that must never cross are redacted at source.
Route
The request goes only to an approved destination: external model, private deployment, or a local model with no egress.
Respond
The downstream answer returns through the boundary, carrying tokens instead of your originals.
Inspect
The answer is checked before anything is restored: an unknown token, a malformed one, or a sensitive value the model produced on its own stops the crossing.
Restore
Where policy permits, tokenised values are restored locally, inside your environment, after the crossing.
Audit
Each decision lands in a detailed audit trail: what was found, what policy did, where the request went, what came back.
BoundSense-authorised nodes
Planned
The same control plane, placed in each customer-controlled environment rather than only in front of one.
What may enter
Which evaluation package or query is allowed to reach this environment at all.
What may execute
Which candidate model may run locally, from the set already permitted for the task.
What may participate
Which local dataset or source is eligible for this specific run.
What may leave
Which metrics, excerpts, structured facts or local answers policy authorises out of the node.
What record remains
The crossing record each node keeps locally, and the identity it contributes to the combined record.
A node is a placement of BoundSense, not a separate product. Nothing here adds a product family or a navigation entry.
A first conversation
Start with
one workflow.
Tell us the task and where the information must stay. We'll show a relevant example and discuss whether a scoped pilot makes sense.
Request a demopilot@kordane.ai