Artificial intelligence could help fight financial fraud

Financial institutions (FI's) face many challenges, but one of the biggest of all is fighting fraud. With so much money and valuable data at stake, the financial system will always be a target for criminals and FI's must shoulder a lot of the responsibility for security.

One of the most effective methods of keeping fraudsters at bay is staying up to date with the latest technologies and there are few areas of technological innovation more exciting than artificial intelligence (AI). It's true that we are still a long way off developing true AI, but the developments that have occurred in this field are already offering great potential for FI's.

Fighting fraud while serving the customer

How do you ensure that your security safeguards and authentication processes are rigorous enough to combat fraud, while delivering the speed and reliability modern-day consumers expect? That's a big question for FI's today, particularly on the payments front.

AI could come in useful here, as emphasized by MasterCard's recent announcement that it was rolling out a new solution to improve the accuracy of real-time approvals and reduce false declines. This could help to address a major source of frustration for consumers: when a perfectly legitimate transaction is rejected.

The launch of Decision Intelligence marks the first deployment of AI on a global scale across the MasterCard network. The system goes further than traditional decision-scoring products, taking a broader approach to assessing and scoring each transaction and learning from previous activity.

Al Pascual, senior vice-president, research director and head of fraud and security at Javelin Strategy & Research, said: "We estimate that in the US alone, the value of false declines is more than 13 times the total amount lost to actual card fraud. Applying machine learning to decision-scoring is a new way of creating a positive consumer experience, while also minimizing fraud."

Machine learning is becoming an increasingly important part of many fraud detection and prevention solutions, which can deliver intelligent, dynamic protection against various threats across all channels, from the ATM to mobile.

One of Australia's biggest banks appears to be getting onboard with AI and machine learning as it looks for new ways to tackle fraud and cybercrime. David Whiteing, chief information officer at the Commonwealth Bank of Australia, told the Australian Financial Review that tools like these can prove invaluable for analyzing large amounts of data.

"When you have a large data set and people can't see the signal for the noise, machine learning capability will sift out and provide you with the top five things that are worthy of further investigation," he said.

Are we getting a bit too excited about AI?

There's no doubt that AI is a topic of much debate and excitement in the financial services industry right now, but do we need to look beyond the hype and take a more pragmatic view of this concept and what it's capable of?

Firstly, it's important to remember that AI and machine learning technologies are still very much in their infancy, with many more years of progress and development required before they are able to realize their potential. For the time being, AI can only complement and improve our existing, human-led processes in the financial services industry, not replace them altogether.

It should also be acknowledged that there have been questions raised about the very idea of AI - whether it's a good thing for the human race or a cause for concern.

As far as financial services is concerned, machine learning and intelligent systems are already a part of the industry, encompassing everything from virtual customer assistants to complex functions that can find patterns in unstructured data. The question now is just how important these technologies will become, particularly in the ongoing fight against fraud.

Dena Hamilton is GM & Director of Enterprise Fraud & Security Solution Management at NCR Corporation.