At the Bank Automation Summit, Informed’s Director of Auto Lending Strategy, Jessica Gonzalez, joined Kevin Faragher, Senior Director of Product and Strategy at Ally Financial on a panel moderated by Whitney McDonald, Deputy Editor Of Bank Automation News.
Here is part of the discussion.
Whitney – What types of fraud are rising? Jessica can you share some figures?
Jessica – Fraud is a hot topic. Car buyers are using a digital interface to purchase and finance cars, so in auto lending, we’re seeing $4.7 billion losses. Informed’s detected fraud average is 2.25% across all of our lenders. Having a digital presence actually increases fraud by .08% – fraudsters are getting more sophisticated and they’re using digital platforms to enable them.
So we’re making sure fraud is contained. Law enforcement is focused on identity theft, because it’s easily punishable and a “hot crime.” We’re focused on paystub fraud because it correlates to consumers paying back their loans. Instead of focusing on identification or KYC, we’re making sure we can calculate consumer income.
Whitney – You’ve talked about paystub fraud and you just released a bulletin. Can you share more about what you’re seeing?
Jessica – The fraud rate across our lenders is ~2.25%. In digital, we see 35% more fraud. A digital retailer is 10 times more likely to see fraudulent pay stubs and documentation across mortgage and lending. In looking at trends, we’re comparing it against that average 2.25%. It may not sound like a big deal, but it’s worth billions. The key is not only having data to track fraud, it’s making sure you recognize trends.
As Kevin said, it’s difficult to manually track trends. Analysts review documents – they see tons of documents daily. They can’t connect all of those data points to uncover trends. When I was at the bank, we saw a telephone bill with a different name and address, but the same telephone number as someone else, and it took almost six months to identify. Having real time, automated transaction analysis is imperative to equip your fraud team and the broader industry by sharing data resources.
AI can take those millions of transactions and highlight trends. So not only having the data but using and analyzing it properly is key.
Whitney – Jessica told us what she sees. Now, Kevin, with Ally – Can you share recent increases in fraudulent activity that you’re seeing?
Kevin – You think about how fraud used to be. Someone stole somebody’s mail, got a fake ID and bought a car. A smart underwriter might recognize that this guy has a credit bureau note in California and they’re applying for a loan in Detroit, which didn’t make sense. But today, everything’s fast. Speed is one of the integral business value propositions.
This fits digital well because fraudsters take advantage, trying to be faster. One of the biggest types of fraud we’re seeing is fraud where folks are partially or completely making up a credit profile designed to get through our underwriting systems. I recently saw an example where somebody had their credit score improved with the model trade line which made the deal score better.
So we review all the data and do a simulation. We have people looking at them, but they’re really hard to spot. When the deal comes through with a synthetic ID you still have to support the identity. That’s where having the ability to have AI capture the bad paystub and flag it for our people is really valuable.
Whitney – You both mentioned how Ally works with informed.IQ to flag fraudulent transactions. Jessica, can you talk through how banks can leverage this technology?
Jessica – Informed automatically detects fraud on paystubs, which is one of the first entry points into the lending process.. So it’s imperative to understand that we think of fraudsters as really high tech and while that can be true, it’s also everyday people faced with a barrier to entry. If you only focus on non- documentary verifications you might run into a lot of synthetic IDs. If you focus on KYC and identity fraud but don’t consume digital documents there is a limitation to how much automatic detection you can enable.
If you’ve received a flat image, just a document image from an email or fax, image quality is an issue. So if you get a fax or a picture of a picture, it’s difficult to know if it’s fraudulent. AI can focus on the ID, but if it’s a flat picture you’ll only succeed 10 to 20% of the time. Most lenders still rely on paper so we’re focused on where we can make significant impact – where we have high confidence we’re uncovering fraud. Relying on Informed’s paystub fraud measure is a good indicator for lenders to make sure they’re identifying not just KYC but also enhanced fraud. Maybe somebody can’t see fraud because it’s a lot easier to get a fake paystub than a fake ID and since there is more focus on KYC and ID verification, more paystub fraud is likely to occur.
Making sure lenders can open accounts and offer a seamless experience for consumers to upload documents is important. If you have those checks at the front end, you can reduce fraud significantly. Making sure you’re checking for fraud at the beginning of your waterfall is critical. Poor image quality correlates to poor performance within the loan portfolio. If you have people that can and will repay loans but cannot provide supporting documentation they most likely will try synthetic ID or a CPN but when we see actual paystub fraud they’re more likely to default. They just don’t have the means to make those payments.
Article: Automation to detect and stop fraudulent transactions