The team reviews exceptions, not PDF stacks.
Fraud signals in claims with a ready dossier, DMHO tampering detected in seconds, and PJ seller onboarding with UBO and restrictive lists. Analysis goes straight to the critical case.
Fraudulent claims hide in volume.
Three pain points that stall insurance claim review and PJ onboarding.
Reading document by document does not scale.
The team receives a PDF stack: report, invoice, DMHO, receipt. Reading each one manually to find tampering is slow and exhausting, and the fraudulent claim slips through mixed with legitimate ones that need fast payment.
Altered dates and amounts pass the human eye.
Edited hospital medical document, document resubmission, altered amount: the fraud is subtle and repeated. Without tampering detection with an audit trail, a PDF edit becomes a paid claim and recurring loss.
PJ seller onboards without UBO or list check.
Onboarding a legal entity looking only at CNPJ ignores shareholders, ultimate beneficial owner, and exposure in restrictive lists. The marketplace or insurer accepts a structure carrying risk no one mapped.
Everything in a single call.
Public and private sources queried in parallel, normalized and weighted by the use-case matrix.
One DMHO. Signal in seconds.
The engine reads the hospital medical document, detects date and amount tampering, cross-references resubmission and provider recurrence, and returns the case to the team with a finding and an auditable trail.
Recommendation: Send to review team with fraud hypothesis. DMHO with tampering signal on date and amount (96.4% consistency), document resubmitted 3 times, and repeat provider. Amount above category average. Consolidated evidence and auditable trail attached to the finding.
Hospital medical documents with altered dates or amounts are flagged in seconds, with the document region that triggered the signal highlighted for the team to verify.
PJ seller onboarding does not stop at CNPJ. Shareholders, ultimate beneficial owner, and PEP, sanctions, and adverse media lists all feed the same decision.
The critical case arrives with a fraud hypothesis and consolidated evidence. The team reviews the exception with context, and the legitimate claim moves to fast payment.
Claims and onboarding regulators require.
Insurance fraud prevention and onboarding due diligence are requirements from the sector regulator and AML rules. Four instruments underpin the control.
AML/CFT in the insurance sector
Insurers follow a risk-based approach with monitoring and identification of suspicious operations. Claims analysis feeds detection.
Insurance fraud
Fraud to receive insurance indemnification is a crime (art. 171, §2, V of the Penal Code). Tampering detection with an auditable trail supports evidence for investigation and denial.
Suspicious activity reporting
Relevant fraud signals in claims or onboarding may feed a COAF report under art. 11 of Law 9.613/1998. The auditable trail backs the decision to report.
Health data processing
Health data in DMHO is sensitive. Processing has a fraud prevention basis, defined purpose, and retention per policy.
From kickoff to go-live in 4 weeks.
The architecture is multi-tenant. What changes are the document types, tampering thresholds, and each client's review policy.
Discovery & scope
Mapping document types (DMHO, invoice, report), PJ onboarding flow, and cases that go to review.
Calibration
Tuning tampering thresholds, recurrence rules, and PJ onboarding matrix. Validation on historical cases.
Directed pilot
Analysis of a slice of claims and onboardings in parallel. First findings for review team validation.
Go-live
Document analysis in production. PJ onboarding with active UBO. Review team receiving ready findings with auditable trail.
Five modules, one decision engine.
One flow from data intake to continuous monitoring. Explore the other platform modules.
Workflows
Orchestrate data, rules, AI agents and the desk in auditable flows. No deploy for every new rule.
Explore
Artificial Intelligence
Agents read the dossier, vote alongside the analyst and justify every decision.
Explore
Decision Desk
Queue, authority, dossier and committee in one screen. Human and AI decide together.
Explore
Risk Hub
Continuous monitoring of people and companies. Automatic re-analysis when something changes.
Explore
Datahub
+40 integrated bureaus, normalized into typed and auditable variables.
ExploreDecide in seconds.
Start with a meeting.
In 15 minutes we show how VAAS works in your scenario, with your rules, your data, your volume.