Visa has rolled out several AI-powered fraud-detecting tools so far in 2024, the latest of which being its account-to-account (A2A) fraud detection service. This tool, which is currently only available to UK-based banks, uses AI-powered analysis to detect A2A fraud in real time.
The pilot program identified 54% more fraud than the current A2A tools being used by UK banks, proving its strong analytical capabilities.
This new A2A tool will hopefully soon be used in the US, where it will join several other anti-fraud tools released by Visa this year, such as the Visa Deep Authorisation (VDA) tool.
The importance of personal vigilance
While tools such as Visa Protect for A2A Payments help prevent fraudulent activity, the user needs to remain vigilant for scammers and only use trusted sites. Staying clear of untrustworthy sites is of paramount importance to avoid scams altogether.
Sergio Zammit has identified the most trustworthy casino platforms that are currently accepting credit cards for payments and withdrawals. These sites come with secure anti-fraud infrastructure, allowing gamers to play casino games and win bonuses without putting their data at risk.
Besides avoiding untrustworthy websites, users must remain wary of phishing, spoofing, and other common forms of fraud. Always verify a user’s identity before engaging in a financial transaction with them over the internet.
Additionally, financial exchanges should only be completed using secure connections, i.e., on private networks with anti-malware software in place.
This, in addition to AI tools, will ensure maximum data privacy and protection.
Visa Protect for A2A Payments
Following the success of the Visa Protect for A2A Payments pilot program, Visa made this tool available to all UK banks. As of 2023, fraud is the most common type of crime in the UK, with over 3.5 million recorded cases last year. This is despite the fact the UK has amongst the most sophisticated payment systems in the world.
The pilot program, which was carried out in collaboration with the UK’s retail payment operator Pay.UK, witnessed AI tools analyzing billions of bank transactions for a year. The program sought to identify fraudulent transactions that had already been completed but had gone undetected.
Now fully up and running, this tool will work to analyze transactions before they are approved and flag any fraudulent activity detected.
The pilot concluded that Protect for A2A Payments could save the equivalent of $419 million.
AI-powered tools already used by US banks
The A2A fraud detector is not the only AI-powered tool Visa has introduced this year. In March, the company introduced three new fraud protection tools to the Visa Protect suite that are currently used by US financial institutions.
Visa is regularly updating its value-adding services to ensure payment protection in five key categories: Identity, Risk, Open Banking, Issuing, Advisory, and Acceptance.
In the last five years, the company has invested $10 billion towards anti-fraud innovation to provide as high a level of security as possible.
The new tools that were introduced and the old tools that were updated in March include:
Visa Deep Authorization (VDA)
VDA is an AI-powered risk-scoring solution designed to better manage card-not-present (CNP) transactions – as in, payments made remotely. This tool uses petabytes of contextual data, recurrent neural network (RNN), and AI risk models to determine the likelihood of a transaction being fraudulent. It will reflect this in the scoring, which will determine whether the transaction goes ahead or not.
It performs this deep analysis promptly and seamlessly, without disrupting the digital transaction.
Visa Advanced Authorization (VAA) and Visa Risk Manager (VRM)
VAA and VRM provide users with premium fraud risk management. These tools have existed for a while but were made network scheme agnostic this year, which made them more accessible to all.
VAA helps users identify positive and negative transactions carried out across VisaNet using in-flight risk scoring. This score pulls together global insights, machine learning technology, and account history data to provide users with accurate scoring.
VRM, on the other hand, creates and tests custom strategies in real time to identify and flag high-risk wallet provisioning requests and purchases. It also features a case management system that analyzes transactions it deems suspicious.
On top of managing whitelists and blacklists, VRM also produces detailed reports that can be used to analyze transaction histories.
Real-Time Account-to-Account Payment Protection
This tool focuses specifically on immediate payments, such as A2A and peer-to-peer (P2P) transactions. Like VDA, this tool analyzes these immediate payments as they occur and scores them using deep learning detection models. If a transaction scores highly, this tool will prevent the exchange from being finalized.
