Case study · Regulated runtime intake

Regulated intake needs proof—not another unstructured ticket

A design-partner intake queue mixed model outputs with unstructured evidence. When an AI-assisted routing recommendation arrived, reviewers could not trace authority or retain admissible proof—until Recon governed the interaction at the boundary.

Design-partner scenario
Intake queueRouting riskPolicy holdAudit-ready proof

Illustrative strip — not evidence from a live regulated deployment.

01 · Before

What was happening before Recon?

Intake tickets combined model version notes, free-text summaries, and attachments with no durable correlation from recommendation to policy decision. Admissibility conversations stalled because portability and authority context were underspecified.

  • AI routing recommendations arrived without inspectable Policy outcomes
  • Reviewers could not show what changed between intake revisions
  • Audit prep required manual reconstruction across ticket systems

02 · Governed moment

What consequential AI action created risk?

An AI-assisted intake workflow recommended expedited routing for a regulated case—moving it ahead of queue policy without explicit reviewer authority. The recommendation could affect downstream compliance obligations if accepted silently.

  • The model suggested priority routing beyond standard intake Policy
  • Missing authority context made the recommendation hard to defend
  • Expedited routing without proof would fail the next audit sampling

03 · Recon intervention

What did Policy permit, hold, or escalate?

Recon evaluated the governed interaction before routing changed. Policy required senior intake reviewer approval when expedited paths bypass standard queue order—turning an opaque model suggestion into an explicit business control.

  • Policy held the routing recommendation pending senior reviewer approval
  • Recon escalated when delegated authority did not match expedited scope
  • The intake workflow enforced the hold instead of auto-applying model output

04 · Proof retained

What Trust Receipt/Ledger evidence remained afterward?

The Trust Receipt captured the routing recommendation, Policy hold, and authority context. The Ledger preserved durable evidence tied to the same correlationId—giving intake stakeholders an admissible proof spine instead of ticket comments alone.

  • Trust Receipt — inspectable artifact reviewers can attach to intake packets
  • Ledger — retention-friendly record for audit sampling and export workflows
  • One correlationId threads from recommendation → decision → receipt → ledger entry
require_approvalTrust ReceiptSample

Recon requires approval before this action proceeds.

correlationId
tacorr_8f3a2b1c4d5e6f708192a3b4c5d6e7f8
receiptId
e5f6789012345678901234567890abcdef1234567890abcdef1234567890abcdef
decision
require_approval
policyId
policy_senior_intake_approval_gate
actionType
regulated_intake_routing_action

Ledger

The Ledger is your durable evidence spine—every Trust Receipt links back to the same correlationId for retention and audit.

correlationId
tacorr_8f3a2b1c4d5e6f708192a3b4c5d6e7f8
ledger.entryId
f6789012345678901234567890abcdef1234567890abcdef1234567890abcdef12
Confirm sample in Ledger →

05 · What changed

What materially changed for operator or business?

Outcome frame: Audit evidence improved. Intake stakeholders gained repeatable proof for expedited routing decisions. Review conversations shifted from reconstructing tickets to opening attributable Trust Receipt and Ledger evidence.

  • Audit sampling used Ledger exports instead of stitched ticket histories
  • Reviewers spent less time proving authority for expedited paths
  • Intake leadership could defend routing holds with inspectable Policy outcomes

Outcome frame

Audit evidence improved

Design-partner scenario only—illustrative regulated intake narrative with no regulator role claimed, no supervisory endorsement, and no measured production ROI.