Fraud Detection · Risk Management
Fraudsters are using AI to find vulnerabilities, adapt tactics, and scale attacks across more payment channels, while traditional transaction-by-transaction controls struggle to catch coordinated fraud and authorized scam payments early enough.
Reported impact
Mastercard says the approach helps issuers detect more fraud while reducing false declines, identifies merchant and scam risk earlier, and gives financial institutions a chance to intervene before harm occurs, supporting safer digital commerce.
Mastercard describes an in-house fraud prevention approach that uses generative AI and broader AI techniques to connect signals that appear low risk when viewed separately. The implementation ingests payment transaction activity along with contextual inputs including location, purchase frequency, transaction value, spending history, device information, account behavior, merchant onboarding signals, digital footprints, and cyber threat intelligence. AI correlates those signals across accounts, devices, merchants, and transactions to surface subtle anomalies, coordinated fraud networks, and scam manipulation patterns that rules-based checks may miss. The resulting risk insights are delivered to issuers and financial institutions so they can intervene earlier, reduce false declines, identify risky merchants before broader ecosystem exposure, and stop scam payments before money leaves an account. The source does not name specific storage platforms or commercial product brands.
Mastercard combines payments intelligence, cyber threat intelligence, merchant onboarding signals, digital footprints, and transaction, account, behavioral, device, and location data. Its AI analyzes connections across accounts, devices, merchants, and transactions to score fraud and scam risk, then surfaces early warnings to issuers and financial institutions so they can block fraud, detect risky merchants, or intervene before funds leave an account.
"At Mastercard, we use generative AI to analyze complex transaction patterns and identify subtle signals that traditional rules may miss, helping issuers detect more fraud while reducing false declines"
"By analyzing digital footprints, onboarding signals and transaction activity, AI can detect potentially fraudulent or compromised merchants before they create broader ecosystem exposure"
"By bringing together cyber threat intelligence and payments intelligence, AI can connect these early warning signs and help identify where risk is taking shape"
"By connecting behavioral, account, and transaction signals, we are using AI to identify manipulation earlier"
"Last year, scam losses globally were estimated at $442 billion, and generative AI could fuel $40 billion in U.S. fraud losses by 2027"
Review how this reference was scored and trace each normalized claim back to its documented source.
Scoring methodology
Customer Reference Value measures usefulness as a customer reference. Evidence Confidence measures the quality and traceability of available evidence, not independent verification.
Traceability
Field coverage shows which facts each source supports. Dates are shown only when documented by the source.
Related
In-house Sample Sep 23, 2024 Zurich Insurance Group Zurich needed to review very large claims volumes more efficiently, reduce investigator drudgery and false positives, and detect fraudulent claims and emerging fraud patterns quickly enough to protect honest customers… a team of 50 specialists Allegro · Zurich’s internal AI chatbot · Zurich’s proprietary fraud scoring tool In-house Sample Feb 1, 2024 Mastercard Mastercard needed to improve fraud detection for banks and cardholders as fraudsters increasingly exploit technology, while also reducing false positives that incorrectly flag legitimate transactions. 143 billion transactions a year Decision Intelligence · Decision Intelligence Pro In-house Sample Jun 25, 2026 Telefónica Telefónica's compliance function must manage an expanding set of digital risks and increasingly complex international regulations without becoming inefficient, while also improving fraud detection and preserving rights… 