Record generator
Builds structured records from runtime events and control decisions.
LandmarkSignal Audit
Evidence and traceability for AI decisions.
Audit is designed to capture decision context, policy outcomes, model routes, and tool actions so teams can review and explain AI behavior with confidence.
What it does
What Audit captures
- Trace IDs linked to decision and execution context
- Policy verdicts, route decisions, and tool-call summaries
- Actor identity and workflow intent metadata
- Structured records for review, retention, and reporting
Where it sits
Architecture position in the AI stack
Runtime actions
→Requests and approved execution paths
Audit generation
→Traceable record creation and linkage
Evidence management
Retention, export, and governance workflows
Core components
Traceability modules
Evidence vault
Organizes records and references for defensible review workflows.
Export interface
Designed to support downstream governance and assurance tooling.
Threats addressed
Close evidence gaps in AI governance
- Missing evidence during incident or compliance review
- Inability to reconstruct why a decision was approved
- Fragmented logs that reduce audit confidence
Deployment patterns
Designed to support assurance environments
Audit is designed to support self-hosted, private cloud, and hybrid deployments with tenant-scoped evidence handling.
Retention controls and export depth may follow roadmap sequencing by environment.
What it is not
Audit is not a generic log aggregator
Audit focuses on AI decision traceability and control evidence.
It is not intended to replace every centralized logging platform.
Use cases and audience
Who uses Audit
Security, governance, and AI operations teams use Audit to prepare defensible evidence for internal and external review workflows.
Example workflow: investigate a flagged decision path, collect trace-linked artifacts, and assemble an evidence package.
Call to action