Visa saves Kiwi businesses $88 million with AI-powered security
Visa’s artificial intelligence (AI) solution Visa Advanced Authorisation has helped financial institutions to prevent nearly $88 million in fraud from impacting New Zealand businesses in the past year, according to a new statement from the company.
Visa implemented a system that uses neural networks, modelled on the human brain, to power its AI technology that analyses the risk of transactions in real-time to identify and stop fraud.
The AI algorithm assesses more than 500 risk attributes in roughly a millisecond to produce a score of every transaction’s predicted fraud probability.
While fraud rates have remained stable over the past year and globbally near historic lows, Visa’s AI-powered security is increasingly critical as payments continue a rapid shift online, where fraudsters tend to commit most of their crime, the company states.
In fact, NZ Post reported that in 2020, more than two million New Zealanders shopped online, up 9.2% on the prior year, and spent $5.8 billion on online shopping - $1.2 billion more than in 2019.
Visa country manager for New Zealand and South Pacific Anthony Watson says, "As consumer spending continues to move online, so has the focus of fraudsters."
He says, "We are investing more heavily than ever in technology that ensures a safe and secure marketplace - combatting fraud while enabling seamless, genuine transactions.
"This investment, which includes a global Visa team of over 850 cyber specialists, covers systems resilience, cybersecurity tools like tokenisation, AI and blockchain-based solutions.”
One of the top threats to emerge for businesses in New Zealand and globally the past year is enumeration, the criminal practice that involves using automation to test and guess payment credentials such as account numbers, CVV2, and/or expiry dates during online checkout.
To counter this, Visa is leveraging another AI-powered solution, Visa Account Attack Intelligence, which spots patterns in data that are otherwise undetectable by humans.
The technology uses machine learning to identify account testing, analyse the details of the attack, and enable Visa to take action in near real-time.
Watson concludes, "The most fundamental attribute in commerce is trust - if a business loses a customer’s trust, they lose sales.
"The global nature of Visa’s network means we’re able to apply learnings from transactions processed by Visa at merchants in every country and territory we operate in around the world to protect New Zealand businesses."
Visa Advanced Authorisation is designed to provide global issuers with sophisticated in-flight risk scoring, for more targeted and better-informed authorisation decisions.
It provides a real time risk score to help identify good and bad transactions across VisaNet, harnesses Visa’s global data and perspective of fraud, uses a two year data profile of account history, and utilises cloud-based fraud risk models using machine learning technology.