Will AI Make the Financial Industry Smarter?

Ralf Ohlhausen, Business Development Director, PPRO Group 

Every year, citizens of the European Union (EU) make 122 billion digital payments using debit or credit cards, bank-transfer apps, e-wallets, mobile wallets and other payment methods. This number is only expected to increase, but the reality is there is no way the payments industry can process so many transactions so quickly, while keeping fraud and error rates down, without the use of Artificial Intelligence (AI). While AI is already widely used in finance, it now has a bigger role to play outside of traditional financial services.

Why is AI necessary?

By 2020, global merchants are expected to process 726 billion digital payments every year. With this volume, heavily relying on the traditional manual review process for each and every transaction is out of the question, as it places a strain on existing fraud-detection systems.

However, the growing popularity of digital payments provides AI developers with the data and the opportunity they need to train and mature algorithms.

A traditional rule-based fraud-detection system might consider a range of variables, such as location, the type of merchant and the amount being spent. For example, if a user spends more than usual, with an unfamiliar merchant in a previously unvisited location, this would be flagged as a possibly fraudulent transaction. This is why cards are often frozen when too much is spent abroad in a single transaction.

The problem with this current model and rule base is that it’s too rigid to cope with increased volume and complexity. Only 1.49% of all global transactions are fraudulent, but in today’s highly digitally-dependent world, many purchases do not fit into a rigid rule-based model of fraud detection.

What can AI do for us?

Done well, AI will make the payment processing industry more intelligent to reduce risk, offer tailored services to customers and ultimately, cut fraud. AI can now perform a task in less than two minutes that used to take a trader 45[1] minutes. A financial institution, using AI to detect fraud, benefits not only from being able to process transactions in real time – something it could not do manually – but also from the ability to recognise the anomalies to successfully distinguish fraudulent transactions from honest ones.

However, a poorly designed AI could incorrectly categorise customers as high risk, denying them access to financial services. Alternatively, a fraudulent transaction could be incorrectly downgraded as low risk. For example, a payments AI will look at a whole range of factors to 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 risk score.

Beyond fraud detection

While fraud detection is the most common use for AI in finance, it is not its only use. AI can also spot potentially useful or worrying connections and behaviour as part of the know-your-customer (KYC) process, which allows institutions to process larger quantities of data for a range of sources. These include all customer accounts and other products with that institution and soon, with the advent of open banking, other institutions as well.

Credit scoring is another area in which the industry is already using artificial intelligence. Again, the challenge here is to analyse data across millions of accounts to spot patterns which correlate strongly with the risk of fraud. Once AI has been used to derive these models it can then check individual applications and customers against them, going beyond simple models based on a narrow range of factors, such as past spending and expected income.

The use of AI for credit scoring also points to some types of risks, both for institutions and consumers. Unless the algorithms are rigorously tested and weighted to avoid replicating bias that already exists in the data, or introducing new biases by making false correlations, then there is the danger that credit-rating-AIs could unfairly deny some people access to financial services.

With all the benefits AI offers, it is undoubtedly the future of finance. But don’t worry, the robots aren’t coming for our jobs just yet. For the foreseeable future, they’re just going to help financial institutions find new customers, serve them better and make the industry smarter than ever.

Ralf Ohlhausen, Business Development Director, PPRO Group 

Every year, citizens of the European Union (EU) make 122 billion digital payments using debit or credit cards, bank-transfer apps, e-wallets, mobile wallets and other payment methods. This number is only expected to increase, but the reality is there is no way the payments industry can process so many transactions so quickly, while keeping fraud and error rates down, without the use of Artificial Intelligence (AI). While AI is already widely used in finance, it now has a bigger role to play outside of traditional financial services.

Why is AI necessary?

By 2020, global merchants are expected to process 726 billion digital payments every year. With this volume, heavily relying on the traditional manual review process for each and every transaction is out of the question, as it places a strain on existing fraud-detection systems.

However, the growing popularity of digital payments provides AI developers with the data and the opportunity they need to train and mature algorithms.

A traditional rule-based fraud-detection system might consider a range of variables, such as location, the type of merchant and the amount being spent. For example, if a user spends more than usual, with an unfamiliar merchant in a previously unvisited location, this would be flagged as a possibly fraudulent transaction. This is why cards are often frozen when too much is spent abroad in a single transaction.

The problem with this current model and rule base is that it’s too rigid to cope with increased volume and complexity. Only 1.49% of all global transactions are fraudulent, but in today’s highly digitally-dependent world, many purchases do not fit into a rigid rule-based model of fraud detection.

What can AI do for us?

Done well, AI will make the payment processing industry more intelligent to reduce risk, offer tailored services to customers and ultimately, cut fraud. AI can now perform a task in less than two minutes that used to take a trader 45[1] minutes. A financial institution, using AI to detect fraud, benefits not only from being able to process transactions in real time – something it could not do manually – but also from the ability to recognise the anomalies to successfully distinguish fraudulent transactions from honest ones.

However, a poorly designed AI could incorrectly categorise customers as high risk, denying them access to financial services. Alternatively, a fraudulent transaction could be incorrectly downgraded as low risk. For example, a payments AI will look at a whole range of factors to 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 risk score.

Beyond fraud detection

While fraud detection is the most common use for AI in finance, it is not its only use. AI can also spot potentially useful or worrying connections and behaviour as part of the know-your-customer (KYC) process, which allows institutions to process larger quantities of data for a range of sources. These include all customer accounts and other products with that institution and soon, with the advent of open banking, other institutions as well.

Credit scoring is another area in which the industry is already using artificial intelligence. Again, the challenge here is to analyse data across millions of accounts to spot patterns which correlate strongly with the risk of fraud. Once AI has been used to derive these models it can then check individual applications and customers against them, going beyond simple models based on a narrow range of factors, such as past spending and expected income.

The use of AI for credit scoring also points to some types of risks, both for institutions and consumers. Unless the algorithms are rigorously tested and weighted to avoid replicating bias that already exists in the data, or introducing new biases by making false correlations, then there is the danger that credit-rating-AIs could unfairly deny some people access to financial services.

With all the benefits AI offers, it is undoubtedly the future of finance. But don’t worry, the robots aren’t coming for our jobs just yet. For the foreseeable future, they’re just going to help financial institutions find new customers, serve them better and make the industry smarter than ever.