Fraud has long been a major issue for financial services institutions. And as global transactions have increased, the danger has too. Fortunately, artificial intelligence has enormous potential to reduce financial fraud. As automated fraud detection tools get smarter and machine learning becomes more powerful, the outlook should improve exponentially.
In its latest report, security company McAfee estimates that cybercrime currently costs the global economy some $600 billion, or 0.8% of global gross domestic product. One of the most prevalent forms and preventable types of cybercrime is credit card fraud, which is exacerbated by the growth in online transacting. The speed at which financial losses can occur when credit card fraud takes place makes intelligent fraud detection techniques increasingly important.
Because of the availability of large volumes of customer data, together with transactional data that is updated as transactions occur, AI can be used to effectively identify credit card behavior patterns that are irregular for specific customers.
My company created an automated, generalizable predictive algorithm that specializes in matching customers and products. I believe it is possible that a similar model could help in the fight against cybercrime. Cybersecurity companies could focus on implementing deep learning to create user and transaction fingerprints by identifying underlying relationships between data points and reducing them to their core components, which they can then cluster together using mathematical models and (depending on a user cluster) can then monitor behavior patterns in relation to other users in that cluster at any given time.
An added advantage of a more sophisticated model is its potential ability to use a wide variety of data points (like Mastercard has already done) to continually fit different customers and transactions into the best-suited clusters for accurate comparison. Thus, as the life circumstances and spending habits of a customer changes, the model would automatically adjust what it views as potentially fraudulent transactions. This could reduce actual fraudulent transactions and minimize false fraud flags (false positives).
False positives occur regularly with traditional rule-based anti-fraud measures, where the system flags anything that falls outside a given set of parameters. For example, if you are planning a trip abroad and you start buying airline tickets and accommodation, this may trigger a fraud warning. A smarter system as described in the two previous paragraphs, that can better understand the underlying patterns of human behavior, could potentially use the new customer data (your travel purchases) to match you with a different cluster of users (for example, holiday travelers). It can then test your behavior against transactions typical to that of the new cluster of users, holiday travelers in this example, before automatically raising a fraud flag on your account.
This should increase customer satisfaction by limiting the number of times that a customer can’t complete a transaction due an incorrect flagging and reduce the operational overheads of the financial institution, by preventing unnecessary interactions with such customers.
The potential for electronic fraud is getting larger with the increased use of advanced technology and the global nature of many transactions. Add to that the newfound ability of cybercriminals to utilize unregulated cryptocurrency exchanges to cash out the return of their criminal online activities, and it becomes clear that it is imperative to use the most advanced techniques available to fight cybercrime.
A Look Into The Future Of Fraud Detection
Most exciting, for those who hope to reduce fraudulent activity even further, is that we are now seeing a new generation of algorithms that are based on the way people think. These are known as Convolutional Neural Networks and are based on the visual cortex, which is a small segment of cells that are sensitive to specific regions of the visual field in the human body. In effect, these neural networks use images directly as input, functioning in the same manner as the visual cortex. This means that they are able to extract elementary visual features like oriented edges, end-points and corners.
This new development in AI makes algorithms that were already intelligent infinitely smarter. This technology can study the spending data of an individual and be able to determine, based on this information, whether they performed the most recent transaction on their credit card or if someone else was using their credit card data. Significant potential lies in the ability of neural networks to learn relationships from modeled data, as mentioned in this World Academy of Science study. Implementing this type of solution to curb cybercrime, for example, will reduce the economic losses drastically.
Fraud has been taking place throughout human history and has only become more complex and difficult to stop, as technology has advanced. Fortunately, we are now in a position where we are also able to leverage technology — especially the new neural networks — to identify these fraudulent activities and stop them before they cause harm.
Achieving this will reduce banks’ overall costs and improve their reputation with customers, who will likely be more loyal to an institution that better protects their money. And there is even the possibility that banks could channel some of the cost savings they make from reducing fraud back to the customers, in the form of lower transaction fees or reduced interest rates. Ultimately, AI looks likely to create a radical shakeup of the entire banking industry, leading not only to reduced cybercrime but happier clients and greater customer advocacy from them. That truly is a win-win situation.