Microsoft Data Protection & AI Governance
Explore identity, managed endpoints, sensitive information, agent oversight and security operations.
Explore lab →Research into practice · Advancing Trustworthy Artificial Intelligence
Reference Labs
Practical environments for exploring AI governance, data protection, security analytics and explainability.
RESEARCH VERSION v0.3 · 30 OCTOBER 2026Explore the environments
Reference designs for repeatable testing and evidence.
Splunk Security Analytics & Threat Hunting
Collect telemetry, develop detections and investigate suspicious activity with repeatable exercises.
Explore lab →Explainable AI — Python, SHAP & LIME
Compare explanations of model predictions and preserve evidence for informed model review.
Explore lab →EVIDENCE BY DESIGN
Test. Observe. Document.
Record configuration, inputs, expected and observed results, limitations and evidence references.
All exercises are Planned. Implementation and validation have not been asserted.
From research to implementation
Each lab includes a proposed baseline architecture, component descriptions, implementation stages, planned tests and its own source register. More reference labs can be added as the research program develops.