SAN FRANCISCO — Visa is reporting it has expanded its account-to-account fraud prevention technology with a new risk-scoring system designed to help banks identify potentially fraudulent transactions before money leaves a customer’s/member’s account.
The enhanced version of Visa’s A2A Protect uses artificial intelligence to provide real-time risk information on account-to-account payments. The update includes a unified fraud score incorporating technology from Featurespace, the fraud-prevention company Visa acquired.
Visa said the system is designed to give financial institutions clearer signals about potentially fraudulent transactions while reducing unnecessary alerts that can interfere with legitimate payments.
The expansion comes as account-to-account, or A2A, payments continue to grow globally. Visa said A2A transactions are projected to exceed 5.8 trillion by 2028, a 160% increase from 2024.
Visa Says System Provides Immediate Risk Data
A2A Protect uses AI and transfer-learning technology to analyze transactions and provide banks with global risk information, Visa said.
The company said the technology allows financial institutions to begin using fraud-detection intelligence without waiting months for models to learn from their own transaction histories or for other institutions to join a data-sharing consortium.
Banks that opt into network-level intelligence sharing also can incorporate additional signals designed to identify fraud patterns emerging across multiple institutions, according to Visa.
“Fraudsters move fast across payment types, and financial institutions need risk insights just as quickly, without slowing down legitimate payments,” James Mirfin, Visa’s head of Risk and Security Solutions, said in a statement.
Mirfin said combining Visa’s payments-network data with Featurespace technology is intended to help financial institutions identify more fraudulent activity earlier.
Visa Reports 75% Increase in Fraud Detection
For participating financial institutions, Visa said A2A Protect can identify emerging scam hotspots and coordinated fraudulent activity that may be difficult for an individual bank to detect using only its own transaction data.
Those signals are intended to give banks an earlier view of potential threats and help identify scams before transactions are authorized, Visa said.
The company said A2A Protect has increased fraud detection by 75% during the first six months after deployment. Visa did not provide additional details in its announcement about the institutions involved in that measurement or the methodology used to calculate the improvement.
Connects With Existing Tech
The system connects with a financial institution’s existing technology through a single application programming interface, or API, which Visa said can reduce implementation time and complexity.
Each fraud alert also includes a plain-language explanation of why a transaction was flagged. Visa said the feature is intended to help fraud teams evaluate suspicious payments more quickly while limiting disruptions for legitimate customers.
Visa operates a global payments network connecting consumers, merchants, financial institutions and governments in more than 200 countries and territories.
