Insights
Field notes on evidence, control design, and agentic compliance.
Practical writing for compliance teams evaluating how agentic systems prepare compliance work, preserve proof, and keep accountable humans in control.
AI content review: who approved this version?
An AI label does not explain a publication decision. Companies need the asset's history, the applicable rules and an accountable reviewer tied to the version that goes live.
Agentic AIA new signal. A new review. What happens when evidence changes after approval
Approval is a point in time. When a new sanctions signal arrives for an approved customer, a reassessment opens against the approved baseline, the comparison shows what changed, and an analyst decides what the change means. Follow that path in cmpliance.
Agentic AIFrom six documents to a KYC decision: a case walkthrough
Follow a KYC onboarding case through document review, an information request, an MLRO decision and an evidence pack. The important work happens between receiving the documents and deciding what they establish.
Agentic AIFour-eyes review: what happens when the checker challenges the recommendation?
A second reviewer should be able to question a recommendation, see the answer and decide whether it is sufficient. This cmpliance walkthrough follows that exchange into the evidence pack.
Agentic AIIs this company connected to a sanctioned party? Follow the path in cmpliance
A relationship map can point an officer toward an important connection without establishing its legal consequence. This cmpliance walkthrough shows why the type of each relationship, its supporting sources and the limits of the finding must travel together.
Audit ReadinessWhy did we approve this customer? Reconstructing the decision
An approved status tells a reviewer what happened. Explaining why requires the source evidence, human rationale and recorded actions that supported the decision. This cmpliance walkthrough follows that path.
Agentic AIAn adverse-media item lands on your desk. What happens next in cmpliance?
An adverse-media item needs more than a headline. Follow the officer in cmpliance from the source record to an information request and an escalation, with the reasoning kept on the case.
Audit ReadinessCompliance evidence glossary: KYC/AML terms, defined plainly
Plain definitions of the terms used in KYC/AML evidence review, from beneficial ownership and adverse media to maker-checker review, evidence packs and hash-chained audit trails.
Agentic AIChoosing an AI agent for KYC/AML evidence: a buyer's checklist
Ten things to require before an AI agent prepares KYC/AML onboarding evidence for your officers, and how to test each one on your own cases.
Agentic AIKYC evidence review beyond name screening
KYC is not complete when the names clear. A defensible review must establish identity, authority, ownership, purpose, expected activity, evidence quality, and human accountability.
Audit ReadinessAudit-ready evidence packs for compliance decisions
An evidence pack should let another officer reconstruct a compliance decision without relying on the original reviewer's memory. Here is what that requires.
Agentic AISanctions exposure through ownership and control chains
A direct sanctions screen can be correct and still miss the exposure. Learn how to investigate ownership, control, evidence gaps, and regime-specific conclusions.
Agentic AICompliance teams were never meant to gather evidence. They were meant to make decisions.
Compliance teams should make decisions, not chase evidence. cmpliance introduces agentic compliance labor for evidence, documentation, audit trails, and human review.
Stay in the loop
Occasional updates on what we’re building, plus walkthroughs from our work.