---
title: 'Mastercard: Decision Intelligence Pro graph-based transaction risk scoring'
slug: mastercard-decision-intelligence-pro
stable_id: ac275cd6b5259281
company: Mastercard
function_code: financial_crime
pattern_codes:
  - creator_to_judge
evidence_strength: reported
publication_tier: showcase
freshness: watch
reviewed_at: '2026-08-23'
updated_at: '2026-08-22'
source_quality_summary: 1 primary; publication outcomes are reported.
caveat_summary: >-
  Internal modeling, not audited production performance. Company analysis with
  proprietary methods; outcome must be refreshed against production evidence.
collections:
  - regulated-autonomy
bundle_version: 1.0.0
bundle_fingerprint: sha256:c23c6cc2b88153d008ea8fda928f632ce0011fc2d4c5036672a16e5d895bab93
canonical_url: https://brianletort.ai/transformations/mastercard-decision-intelligence-pro
---

# Mastercard: Decision Intelligence Pro graph-based transaction risk scoring

A sub-50-millisecond model can enrich transaction risk before the issuing bank decides approval.

Function: Financial crime. Patterns: Creator to judge. Evidence: reported.

Freshness: watch. Reviewed: 2026-08-23. Updated: 2026-08-22.
Watch status: verify the cited source and deployment condition before reusing this case.

Source quality: 1 primary; publication outcomes are reported.

## Before

1. **Decision Intelligence** — Scores transactions from account, merchant, device, and purchase features. (control: Bank fraud rules and authorization policy.)
2. **Issuing bank** — Approves, declines, or challenges the transaction. (control: Issuer decision rights.)

## After

1. **Decision Intelligence Pro** — Analyzes entity relationships across a much larger graph and improves the risk score in under 50 milliseconds. (control: Real-time bounded scoring.)
2. **Issuing bank** — Uses the score with its own rules to approve, decline, or challenge. (control: Issuer retains authorization.)

## Decision rights

Issuers decide approval, decline, challenge, and customer remediation.

## Exception path

Uncertain or challenged transactions follow issuer verification and fraud-investigation processes.

## Outcomes

- **Fraud-detection model performance** (reported): Existing Decision Intelligence scoring. → Mastercard's initial modeling showed 20% average fraud-detection improvement.. Internal modeling, not audited production performance.
- **False positives** (reported): Existing Decision Intelligence performance. → Mastercard's analysis projected more than 85% reduction.. Company analysis with proprietary methods; outcome must be refreshed against production evidence.

## Executive lesson

High-scale risk AI can qualify as a workflow case while its performance claims remain explicitly company-modeled and refresh-sensitive.

## Anti-pattern

Publishing modeled gains as audited realized results.

## Questions for leaders

- What production cohort verifies the model?
- How do issuers override or tune scores?
- What protected or proxy features are monitored?

## Sources

- [Mastercard supercharges consumer protection with gen AI](https://newsroom.mastercard.com/news/press/2024/february/mastercard-supercharges-consumer-protection-with-gen-ai/) — Mastercard