Forward-looking financial institutions are looking to incorporate the latest technology to stay competitive. But with new technology comes new security concerns. With the introduction of the “digital identity” through IoT, banks are faced with vast amounts of data and no idea how to use it.
Customers demand faster payments but are annoyed with authentication techniques. How can organizations find a balance? The solutions are simple, if you’re willing to make an investment in customer loyalty and operational efficiency. As the experts will tell you, the investment is well worth the reward.
We sat down with our experts, David Stewart, Banking Industry Director for SAS Fraud and Security Intelligence; Ian Holmes, Global Lead for Enterprise Fraud Solutions for SAS Fraud and Security Intelligence; and John Watkins, Advisory Industry Consultant for SAS Fraud and Security Intelligence, to talk about the issues weighing on financial institutions. Here are their thoughts on the most talked about issues in the industry:
Q: Financial institutions seem to have a love/hate relationship with the concept of AI. Why should they be embracing this technology vs. fighting it?
David Stewart: Financial institutions should embrace several sub-disciplines of AI in combatting fraud and financial crimes. These techniques will allow institutions to more effectively authenticate customers, improve customer experience, and reduce the cost of maintaining acceptable levels of fraud risk, particularly in digital channels.
Machine learning is a proven method that automates some of the supervised learning techniques in areas of fraud, with good training data on fraud events. We’re now seeing these approaches like decision trees, neural networks and GBM models being applied in anti-money laundering to predict “productive events.” Some of the advancements in linguistic analysis and contextual text analytics are proving helpful to automate tasks that have been historically performed manually. Any time you can reduce false positives by 50-70% with automated machine learning strategies, you’re freeing up precious human resources that can focus on more complex and subjective investigations.
Some will experiment with AI, but this year will set a strong foundation for more widespread adoption of AI in business operations in coming years.
Download free white paper: How to Succeed with Fraud Analytics
Any large organization has vast resources of data that could power an analytics-driven defense against fraud. Read this paper to learn the three pillars of analytics success and six ways to make a solid business case for funding a new or upgraded fraud analytics program.
Q: The payments industry has been, and will continue to be, one of the most dynamic areas of innovation in the banking industry. What impacts will we likely see from the demand for faster payments?
John Watkins: Real-time transactions offer significant flexibility and convenience to both banks and consumers. As more and more countries look to implement instant payments, mitigating fraud is a key consideration for banks. Real-time fraud decisioning is required in digital channels, as well in opening new accounts. A combination of customer, device, and session behavioral analysis is required to prevent fraud losses. And with PSD2 on the horizon, the potential use-cases and business models will increase exponentially.
Ian Holmes: Faster payments, however, do present challenges for financial institutions. They’ll drive high velocity fraud attacks, leaving traditional systems unable to cope with the new types of fraud that will occur at a rapid pace. Most systems have only milliseconds to assess risk and identify potential suspicious activity -- limiting an institution’s ability to “claw back” fraudulent payments. Without a strong real-time system, financial institutions could see a spike in false positives from poor customer recognition.
We’ll also see new industry players (fintechs, PSD2 third-party payment providers and other intermediaries) will add more complexity to core business operations (fraud, credit risk, marketing, etc.).
Any time you can reduce false positives by 50-70% with automated machine learning strategies, you’re freeing up precious human resources that can focus on more complex and subjective investigations. David Stewart Banking Industry Director for Fraud and Security Intelligence SAS
Q: We all agree that customer experience is a top concern. How can the advancements in advanced analytics make this easier?
Ian Holmes: The digitization of banking has allowed institutions to gather valuable data about their customers. That includes what times of day customers typically access online accounts, devices used to gain access, and a general range of transaction types. With machine learning informed by all this transaction data, systems can learn when a transaction falls outside the norm and alert the customer that something may be wrong.
John Watkins: This information can also help to authenticate a customer before they even enter their information. The ability to authenticate without additional customer information protects both the consumer and the company. Organizations are always seeking that optimal balance between reducing the false positives that can damage customer relationships, and the false negatives that can lead to financial loss for the institution. That requires analytics, the ability to detect anomalies that represent potential red flags – at the speed of now – in an ever-changing fraud environment.
Q: With the introduction of multiple digital channels to conduct transactions, there’s even more data out there on a customer. With all the data out there, what do financial institutions need to do to prepare for digital fraud prevention?
David Stewart: Financial institutions understand that digital fraud prevention is as much about digital experience as anything. The industry has moved from payment authorization to identity verification. To understand if a session or digital interaction is authentic, there’s a lot more data required that may include geolocation, session behavior, and device profiling in addition to other data from the merchant or issuer. Ideally, we want to score the person so that legitimate customers have an easy time in the app, online, or at point of sale. Orchestrating all the interactions between external data, consortium data, bureau data, etc. with customer profiles and delivering a sub-second decision is no easy task.
Institutions will need to have a scalable digital hub as we move into ambient banking in the 2020’s.
- Detect and prevent banking application fraudCredit fraud often starts with a falsified application. That’s why it’s important to use analytics starting at the entrance point. Learn how analytics and machine learning can detect fraud at the point of application by recognizing the biggest challenge – synthetic identities.
- Health care cost containment through big data analyticsHealth insurers are plagued by fraud, waste and abuse. For health care cost containment, an enterprise approach to payment integrity using data management and analytics can help. With this approach, payers can detect and prevent fraud; influence provider, employee and patient behavior; and substantially reduce costs.
- 10 tips for Life Insurance Companies and AMLLife insurance companies have AML challenges that are both simple and complex. The challenges are simple as there are less products and channels available compared to banks. Here are 10 tips to consider to stay ahead of the international crime syndicates.