DECISION PROOF LAB
A model changed. Find the past decisions that may need review.
Use one continuous synthetic payment story to find affected decisions, understand one decision, test what can be proven, inspect a system disagreement, and verify a bounded proof snapshot.
The public Lab is frontend-only, synthetic, read-only and off-path. It is designed to make the product experience tangible without exposing the private backend or using customer data.
PUBLIC PRODUCT JOURNEY
01 ChallengeChoose what changed02 ImpactFind affected decisions03 InspectOpen proof, gaps and conflicts
THREE PUBLIC EXPERIENCES
One interactive investigation, one deep decision inspector and one adjacent applicability example.
The Lab prioritizes depth over a large catalogue of shallow demos. Future domains remain clearly separated from what visitors can actually test today.
Model Change Impact Investigation
Choose a defective model, changed rule or source-mapping issue. Filter directly impacted, potentially impacted and unaffected decisions; inspect why each decision is included; then open its proof state.
Start investigationPayment Decision Proof Twin
Inspect retained payment, authentication, risk, configuration, human-action and outcome context for one decision, with chronology and limitations visible.
Inspect decisionAML Alert Review Proof
See how the same Decision Proof pattern applies to a different source workflow without turning RecordArc into an AML monitoring, case or filing system.
Explore AML exampleIMAGINE THIS HAPPENED
A bank discovers that Risk Model 2.3 may require review after payment decisions have already been made.
Your job is to find which past decisions may be affected, understand one decision, test what the retained evidence can prove, expose disagreements, and identify what needs human review.
Start investigationYOUR INVESTIGATION
Follow the question in front of you; inspect the technical evidence when you need it.
The existing expert workspace remains directly usable. This visible sequence explains where you are, why the current view matters, and what to inspect next.
- 01
What changed?
Start with the challengeA model, rule, policy, configuration or source mapping changes or is challenged after decisions have already been made.
- 02
Find affected decisions
Decision Impact RadarFind the decisions directly affected, potentially affected or supported as unaffected by the retained evidence.
- 03
Understand one decision
Decision Proof TwinSee the retained evidence, versions, human actions and outcomes around one decision.
- 04
Test what can be proven
Challenge ReadinessSee which questions are answered, partial, blocked, conflicting or not applicable — and why.
- 05
Find system disagreements
Cross-System DivergenceExpose contradictory retained outcomes or evidence without deciding which source is correct.
- 06
Review and carry the proof
Human Review Cohort · Portable Proof CapsuleIdentify the decisions that need authorized investigation and provide a bounded evidence snapshot with limitations and integrity information.
Loading synthetic Decision Impact workspace…
WHAT YOU CAN ACTUALLY DO
The flagship Lab behaves like a small product, not a static marketing mockup.
Every public control is intentionally tied to a meaningful decision-proof task.
Switch between a model defect, rule change and mapping issue.
Compare direct, potential and unaffected decisions.
Open evidence, governing versions, human actions and recorded outcomes.
Open answered, partial, blocked and conflicting proof questions.
Verify a synthetic proof capsule, then simulate modified evidence and see it fail.
FUTURE APPLICABILITY
Other regulated decisions may reuse the pattern after the first wedge is validated.
KYC/KYB, sanctions, credit and other high-impact decisions remain future applicability — not equal public product experiences today.
KYC / KYB
Future applicability
Sanctions review
Future applicability
Credit decisions
Future applicability
Insurance decisions
Future applicability
AI / agent governance
Future applicability
Other regulated decisions
Future applicability
START WITH THE FLAGSHIP
Challenge the synthetic decision population yourself.
Use the interactive Radar first, then drill into an individual payment decision if you want the deeper proof chronology.