Will AI make the financial industry smarter?

When you hear the words “artificial intelligence”, what do you think of? A machine that thinks, communicates, and behaves like a human? This kind of general AI still exists only in science fiction. But other, contextually specific types of AI not only exist, they already help power the payments industry.

What do we mean by AI?

The AI we have now is known as “weak AI”. It is generally focused on one task which it repeats infinitely, maximising its efficiency and effectiveness as it goes. No weak AI can truly replace a skilled, multi-tasking human worker.

The most common type of AI is one based on machine learning (ML). Programmers teach an ML-based AI to train itself by giving it training data, along with the desired outputs for that data. The AI then processes the data in various ways until it achieves the correct outputs and its programmers are satisfied it can work with future data in the same way and achieve the desired results.

Another well-known type of AI currently in use is deep learning, a subset of machine learning. This is based on something called a neural network. Designed to mimic the brain, a neural network consists of layers of nodes, with each node standing in for a neuron. Each node assigns a probability to the input, based on how likely that input is to generate one of the desired outputs.

For instance, if you were designing a deep learning system to recognize known faces, the first layer might look for patterns that resemble particular features. If it found something that looked enough like a nose, it would pass this information on to the second layer of neurons. These might then check the identified features against a known set of values. Do I know this nose? Yes, this looks like Bob’s nose. If the AI makes enough of such matches, Bob is cleared to enter the building.

There are other models for AI. In the neuroevolution model, for instance, a number of weak AIs are designed to solve a specific problem. The one that gets closest is used as the basis for the next generation of AIs. This continues until you have an AI that can do the job you need it to do.

What can AI do for us?

AI is already being used to automate repetitive tasks. UBS, for instance, recently automated its post-trade allocation requests. An AI can now perform in less than two minutes tasks that used to take a trader 45 . Similarly, Google has developed a medical AI which is able to accurately scan patient imaging data for cancer, a job that used to take doctors five or six hours. In the payments sector, AI already powers many fraud detection and AML systems. And, as the volume and complexity of payments increases with the adoption of open banking, the industry will come to rely more and more on AI-systems to detect fraud.

A 2017 study of merchants in North America found that 79% used manual reviews to determine whether at least some anomalistic transactions were likely to be fraudulent. On average, merchants manually reviewed 25% of all transactions. According to another study, 52% of transactions flagged as fraudulent were false positives.

Even today, this kind of manual review process is bad for business. But by 2020, global merchants are expected to be processing 726 billion digital payments every year. With those kinds of volumes, no one can afford to rely heavily on manual reviews.

Using AI, usually machine learning, can significantly improve fraud detection and reduce false positives. For example, a payments AI might look at a whole range of factors and assign a risk score to each. A merchant with a good track record might have a low-risk score, say 15%, but an unfamiliar IP address, time zone or location might attract higher risk scores.

This process can be repeated for hundreds of factors, with the final average score determining whether the transaction passes the merchant’s threshold for being flagged up as fraudulent. It’s possible in this way to analyse vast quantities of data to build a much more sophisticated picture of what “normal” looks like. Because it’s not limited to working within set rules, an AI can analyse this larger body of data to look for unexpected commonalities between both fraudulent and non-fraudulent transactions.

Nor is fraud detection the only way in which heavier investment in AI can benefit the payments industry. The same versatility in pattern detection can be applied in areas such as the know-your-customer (KYC) process, credit scoring, and even marketing — using AI to spot a customer’s present and future needs, and predict which products will be most relevant to them.

AI is the future of finance. But don’t worry, the robots aren’t yet coming for your jobs. For the foreseeable future, they’re just going to help you find new customers and serve them better.

About Ralf Ohlhausen:

Ralf Ohlhausen, MSc in mathematics and Master of Telecommunications Business, has over 25 years’ experience in ecommerce, financial services, mobile telecommunications and IT. Before joining PPRO Group, he was President Europe at SafetyPay. Other management positions on his international career path took him to Digicel, O2, British Telecom and Mannesmann-Kienzle. At PPRO, Ralf is responsible for increasing PPRO’s global reach, focusing in particular on the addition of new payment choices to the company’s portfolio.

About PPRO Group:

Cross-border e-payment specialist, PPRO Group, (PPRO) removes the complexity of international ecommerce payments by acquiring, collecting and processing an extensive range of alternative payments methods for Payment Service Providers (PSPs) under one contract, through one platform and one single integration. PPRO supports international payment methods across more than 100 countries, allowing PSPs to expand their merchants’ ecommerce reach, arrange hassle-free collection and achieve higher conversion rates.