Data protector
Security agent
Your team shares information with AI tools every day. Data Protector makes sure sensitive content is caught before it leaves your environment.
Keep sensitive data out of AI interactions
You define what counts as sensitive. Personal identifiers, financial records, credentials, health data, or categories specific to your organisation. Data Protector recognises these patterns across every AI interaction and gives you control over what happens next.
Configure detection categories that match your data sensitivity requirements
Choose per category whether to block, redact, or warn
Escalate uncertain cases to a human reviewer when context is needed
How Data Protector works
Data Protector scans content wherever your team uses AI. Every prompt, paste, or AI-triggered action is evaluated before data reaches the external service.
1
A team member writes a prompt, pastes data, or triggers an AI action in the browser or a connected system.2
The agent analyses the content against your configured categories and detection patterns.3
Based on your rules, the interaction is allowed, blocked, redacted, or routed to a human reviewer.4
Each interaction is logged to surface patterns and help you refine your configuration over time.
FAQ
Answers to your questions
Common questions about how the Data protector works and what to expect when your team starts using it. If you do not find what you are looking for, dont hesitate to reach out.
Get in touchIt does not store prompts, messages, or generated output. Content is scanned in memory and discarded once the check completes. Only metadata about the interaction is retained for reporting.