Feeds

Key considerations for managing digital fraud at speed and scale

According to Feedzai CEO Nuno Sebastiao, banks can deliver a frictionless customer experience and react to fraud faster by leveraging the data across their different digital channels

This editorial was first published in our Web Fraud Prevention and Online Authentication Market Guide 2017/2018. The Guide is a complete overview of the fraud management, digital identity verification and authentication ecosystem provided by thought leaders in the industry from leading solution providers (both established and new players) to associations and experts.

Consumers are increasingly seeking digital channels to transact, whether for shopping or opening a bank account. Open banking initiatives like PSD2 in Europe have been creating a competitive environment for banks, in which data has become the new currency. According to PricewaterhouseCoopers, in 2015, the year of the PSD2 release, the UK saw more electronic payments than cash payments for the first time. And in 2018, the year of its implementation, 20% of online transactions are likely to be made with mobile devices.

With the rise of digital commerce and the proliferation of new channels and payment types, data can be a catalyst for improving customer experience. On the other hand, data also creates an exposure to vulnerabilities like new fraud patterns, massive fraud attacks, and data breaches. The result of our growing digital economy is a steady increase of fraud across the globe. Global card fraud losses have nearly quadrupled since 2010, from USD 7 billion in 2010 to USD 27 billion in 2017, according to The Nilson Report.

One characteristic of today’s fraud is the high financial loss. Beneath the money loss, there’s everything else, from law penalties to operational disruption. Furthermore, every breach opens the door to new fraud activities, like account takeovers and massive fraud attacks.

But the costliest consequence of these crimes is that people get hurt. When someone has his identity or credit card information stolen, society suffers. Managing risk is rooted in a goal deeper than saving money. The goal is to make society safer.

To achieve this goal, organisations must deal with a key characteristic of today’s fraud: its high velocity. Criminals deploy speed as a tool and leverage the most advanced technology to launch rapid attacks. For example, at Feedzai, we’ve discovered fraudsters using bots that fill forms five times faster than humans.

Banks have come to conclude that they need more sophisticated tools in order to detect fraud at scale and in sub-millisecond transaction time. Banks are turning to AI systems to help them navigate a complex set of broader goals: mitigating risk, remaining competitive, and offering cutting-edge customer experience. Here are three considerations to be taken in the pursuit of that machine learning system that can enable banks to stay ahead of new and evolving fraud.

Consideration #1: Refocus on the customer with a complete view

Because banks today are product-centric, rather than customer-centric, they make decisions in silos and are vulnerable to attacks across multiple channels for the same account. How can a bank know whether a customer defaulting on a credit card bill is a risky customer for a home mortgage?

Machine learning can break down data silos by performing omnichannel aggregation and omnidata integration. The result is a 360 degrees view of transactions right as they happen.

To build up a complete customer view at speed and scale, a machine learning system needs to be data agnostic, to be conceived for the purpose of extracting and loading all kinds of data, whether they are within the bank’s own system or are augmented from external sources.

Consideration #2: React to fraud faster

With digital transformation comes the expectation of immediacy. Customers want decisions made at the speed of transactions. The increase of immediacy trends in payments, such as Amazon 1-click ordering, are only reducing the amount of time it takes for customers to transact.

To manage the risk associated with immediate transactions, banks need a system that can accelerate the machine learning process and lead fraud analysts and data scientists to fraud drivers more quickly. The benefits of speeding up the machine learning process are explored more deeply in the report Improving Fraud Detection by Speeding Up Machine Learning.

To enable organisations to react to fraud faster, a system must be also architected for the rapid deployment of new models. Furthermore, as analysts review transactions and label them as fraud or not, the platform should integrate with those decisions and automatically learn from them to become better at recognising future patterns. 

Consideration #3: Pursue explainability

As more and more banks are turning to machine learning to make good decisions, they’re realising the need for explanations too. Transparency and interpretability in a machine learning system have two important benefits.
First, a system with interpretable reasoning lets organisations audit the machine and provide trails of explanations for compliance. Second, better explanations means banks and merchants can drive greater engagement and deliver greater customer experience.

Today explainability exists in the form of whitebox processing, which adds human-readability to the underlying machine logic and communicates factors behind its decisions to the human analyst. What does the next stage of explainability look like?

What’s next for machine learning?

My colleague, Pedro Bizarro, Feedzai co-founder and CSO, has said that the critical need for explainability calls for a machine that can link patterns with increasing complexity, and explain even more underlying connections to “what’s actually going on.”

The maturation of explainability will expose humans to more sensitive information, which raises questions around ethics. To reduce bias in AI, and to protect the integrity of enterprise data and the privacy of machine insights, Bizarro has developed an internal AI Code of Ethics, and he’s asking AI practitioners to help shape the future of ethical machine learning. He explores this future in the report What’s Next for Machine Learning: Ethics and Explainability in AI for Fraud

About Nuno Sebastiao

Nuno is co-founder and CEO of Feedzai, an agile machine learning platform for risk management. Previously, Nuno led the development of the European Space Agency Satellite Simulation Infrastructure, contributing to the Rosetta space probe. Today, Nuno and his team of top data scientists and experts in payments technology are working to achieve a singular mission: to make commerce safe

About Feedzai

Feedzai is coding the future of our digital economy with the most advanced risk management platform, powered by big data and artificial intelligence. Some of the world’s largest organisations use Feedzai’s machine learning technology to manage the risk associated with banking and shopping, whether it’s in person, online, or via mobile devices.

Source link