What is MultiLayer Data Shield?
4 protection layers, technically explained.
PII Detection
Real-time scanning of all inputs for personal data: names, emails, phone numbers, credit cards, IBAN, addresses, social security numbers and more.
Document Analysis
Uploaded files are checked for sensitive content before processing. Admins control allowed file types per user group.
Prompt Filtering
Configurable rules prevent confidential information from entering AI prompts. NSFW filter and content policies per group.
Audit & Logging
Every interaction is fully logged: who, when, which model, what data detected, what measures taken.
Frequently asked questions about the MultiLayer Data Shield
What is the MultiLayer Data Shield?+
The MultiLayer Data Shield (MLDS) is HOVIGuard's protection architecture between users and AI models. It combines four layers: PII detection, document analysis, prompt filtering and audit/logging. Individual layers are configurable through the admin interface.
Which layers does the MultiLayer Data Shield include?+
Layer 1 inspects inputs for personal data (PII). Layer 2 analyses uploaded documents. Layer 3 filters prompts based on configurable content and security rules. Layer 4 logs each interaction. Layers can be configured per group.
Which inputs does the MLDS inspect?+
Typically chat prompts, uploaded files (subject to allowed file type configuration) and model responses. Which fields are inspected in detail depends on tenant configuration and the respective group.
Does the MLDS provide complete protection against data leakage?+
No. Full protection cannot be technically guaranteed. The MLDS reduces risks through multiple layers and delivers an audit trail. Effectiveness additionally depends on configuration, internal policy and user training.
How configurable are the layers?+
PII categories, allowed file types, content filters (e.g. NSFW) and audit detail level can be configured per user group through the admin dashboard. The platform help documents the available options.