Taking our best-in-class machine learning platform to LendIt Fintech USA
Simility spent a few days last week at Lendit Fintech USA, one of the key tech innovation events on the financial services calendar. We had a productive time reaching out to industry practitioners, listening to their pain points and explaining how our Adaptive Decisioning Platform can help finance risk teams detect fraud more effectively without adding friction.
Financial services under fire
The financial services sector is among the most commonly targeted globally. It’s easy to see why: by hijacking existing customer accounts, fraudulently opening new ones and/or applying for loans, cybercriminals can generate significant revenues. Their job is made easier thanks to the expansion of digital transformation efforts designed to help lenders become more agile and customer-centric. In reality, by offering more online services, banks are also exposing themselves to greater fraud risk.
One rapidly rising type of banking fraud involves fraudulent applications for new loans: for example, applying for new credit cards. Unfortunately, many banks are unable to accurately validate the identity data used on such applications. In many cases, the fraudster uses “synthetic” identities comprised of real and fake information, making scams harder to spot. It’s a trend said to have cost lenders billions, with the FTC claiming it’s the fastest type of identity fraud around.1
We’re also seeing synthetic fraud being taken to whole new levels, with scammers exploiting loopholes such as lenders’ inability to verify multiple loan applications at the same time. This so-called “loan stacking” can help them maximize their ROI — and by the time they’ve defaulted on these loans it’s too late for the financial institution.
The power of machine learning
Lenders are understandably reluctant to add in extra fraud checks which may harm the customer experience and hold up the speedy processing of applications.
The key is to seek out fraud platforms that combine multiple methods to verify identities, drawing data from static and dynamic sources including device, transaction, third-party, session, historical, and many more. By applying machine learning models to this large volume of internal and third-party data, financial institutions can then detect anomalous patterns which may be missed by human eyes. That’s the path to highly effective fraud prevention which doesn’t impact the end user experience.
It’s a message that was well received at LendIt Fintech last week. Machine learning and AI was one of the standout themes of the show. But we believe it’s best used in fraud prevention to support human analysts, not replace them.
Ultimately, the feedback we get from customers and industry experts at events like LendIt Fintech help us continue to develop products, services and features that meet the needs of businesses around the world. But don’t take our word for it: Simility was one of just three providers judged to provide best-in-class fraud and AML tools, according to a new independent report from analyst Aite Group.2 Read what our customers had to say here.
FTC, The changing face of identity theft, https://www.ftc.gov/sites/default/files/documents/public_comments/credit-report-freezes-534030-00033/534030-00033.pdf
Aite Group, based on 14 fraud and AML machine learning platform vendors, March 2019
Latest posts by Vanita Pandey (see all)
- Taking our best-in-class machine learning platform to LendIt Fintech USA - April 18, 2019
- Taking the Adaptive Fraud Detection Message to MRC Vegas 2019 - March 19, 2019
- Adopt Clear-box Machine Learning to Make Explainable Decisions - March 7, 2019