AI workflow map
A clear list of the AI workflows, owners, users, platforms, source classes, and decision points in scope.
Primary offer
Company AI Evidence Audit.
A scoped review that maps AI workflows, source boundaries, exclusions, approval gates, validation needs, and context-preservation gaps before AI access expands.
What gets audited
Not your entire company indiscriminately. The AI evidence boundary.
A strong audit is not a broad scrape or a promise to “tell you everything.” AuditTrace reviews the workflows where AI may touch business context, then documents what should be approved, excluded, validated, and preserved for later review.
Core audit question
Can the organization prove what the AI workflow was allowed to use, what stayed out, who approved it, and what changed after launch?
Deliverables
What the buyer receives.
Approved source register
A record of source classes the AI may use, with owner roles, business purpose, sensitivity, and freshness notes.
Exclusion register
A record of source classes that must stay out, including legal, HR, finance, private inboxes, client-specific folders, and unmanaged exports.
Validation question set
A set of approved-answer tests, excluded-source refusal tests, citation checks, and escalation scenarios.
Evidence manifest
A compact manifest tying the workflow to source boundary, approval status, validation status, and preservation needs.
DecisionCard
A go / limited / blocked / remediation-needed decision artifact for human review.
Offer levels
Start narrow, then expand only when the evidence supports it.
| Package | Best fit | Includes | Anchor |
|---|---|---|---|
| Starter Workflow Audit | One AI use case or first pilot | Workflow declaration, source boundary, exclusions, validation set, first-pass DecisionCard. | $1,500–$2,500 |
| Department AI Evidence Audit | Department or discrete business unit | Two to four workflows, source-boundary matrix, owner map, validation scenarios, flagship pack. | $5,000–$9,500 |
| Company AI Evidence Audit | Multi-team or executive-sponsored rollout | Company-wide AI use map, priority workflow ladder, shared exclusions taxonomy, preservation triggers. | Custom |
What this is not
Clear limits make the service safer to buy.
AuditTrace does not certify every output, replace legal counsel, monitor employees, bulk-ingest uncontrolled data, run covert collection, or enable autonomous actions. The audit creates a reviewable evidence artifact around approved context and human gates.
Best first inquiry
Tell us the AI workflow, platform, source classes it should use, source classes it must not use, owner role, final approver, and what you need to review later.