10 Fraud Detection Questions From Our Recent Webinar

Pindrop’s recent webinar was hosted by Shawn Hall, VP of Product, Research, and Engineering at Pindrop and Ketuman Sardesai, Head of Market Strategy and Intelligence at Pindrop. With a focus on Pindrop’s 2024 Voice Intelligence and Security Report, the webinar covered important questions affecting the industry—and aimed to help you create a strategy to better detect fraud in the future.

1. How many data breaches occurred in 2023?
   Data breaches in 2023, as analyzed in Pindrop’s recent Voice Intelligence and Security report, were 78% higher than the previous year, and over 353 million victims were targeted in 2023. Outdated and old-school technologies are also depleting consumers’ trust. The report explains that passwords are the most used but are the least trusted in security measures. With consumer uncertainty, 93% expect strong security measures in the future.
2. What can you do to combat fraud and improve customer trust?
   You need multi-factor authentication to build trust. You also need technology to fight new generative AI, such as deepfakes. Five to six years ago, it took a fraudster three to six months to switch tactics. “Now, they switch things up weekly. It’s nearly impossible to catch fraud without robust technology to see and identify it before it happens,” says Shawn Hall on the webinar.
3. Does fraud vary by industry?
   Pindrop’s new Voice and Security Intelligence report projects that, by the end of 2024, more than 1 in every 730 calls into call centers will be fraudulent. The current rates vary by vertical, with retailers experiencing the highest fraud rate at 1 in every 264 calls.
   According to an analysis by Pindrop^Ⓡ Labs, 1 in 99 calls at a large U.S. retailer are fraudulent. Shawn says, “In recent years, 2020-2021, the fraud call rate dropped significantly as fraudsters targeted PPP loans and government stimulus money. This took some focus off call centers, but the volume quickly returned to contact centers once those funds dried up and has significantly increased in the past 2 years.”
4. What are fraudsters’ top strategies for gaining access through the IVR?
   Some top fraud strategies, according to Shawn, include:
   • • • Password resets
       • Account/information reconnaissance
       • Changing account information
       • Account takeover
       • Quick access for wiring funds

   The contact center accounts for 62% of all fraud instances, and Shawn thinks this will increase to 70% in 2024. Shawn says, “Fraudsters are leveraging IVRs as a person would utilize a Google search engine to learn more information about account details and also understand an organization’s security framework, which puts them at a great risk without the proper security measures.”

5. What can companies do to protect themselves in the IVR?
   Companies have been trying to put more proactive measures in the IVR, hoping it will drive more customers away and lead them to other self-service technologies and mobile apps. “The problem is these new technologies are equally accessible to fraudsters,” says Shawn. He explains that the key is only to allow sensitive information to be available to low-risk customers and to leverage technology to help determine that through a risk score. This element will enable customers to get routed to self-service only when low risk. The higher the risk score, the more likely the caller would be routed to a specialized contact center agent to handle the call and request.
   
6. How do you know if you’re interacting with a real person or a deepfake?
   Quiet reconnaissance is happening in the call center to learn security patterns through synthetic voice. “Account takeover is the big thing they are utilizing here,” says Shawn. Many fraudsters contact customers directly, dupe them, get them to give up information, and capture their voice in the process. Once they have the voice, they can use text-to-speech to access accounts. One example is how tough it was to spot the difference in the Joe Biden deepfake in the primary, which told voters not to vote in New Hampshire. Fraudsters are capitalizing on various channels to take over accounts and gain information as well. 
7. What trends do you expect to see in 2024?
   “We’ll have the 3,000 range and maybe get above 4,000 breaches this year. Large financial organizations see more like 1 in 500 calls, and retailers see 1 in every 50,” Shawn predicts for 2024. Some customers are already seeing this today. Companies want to expand self-service and into other areas, but how can fraud occurrences not increase with the emergence of deep AI and deepfakes?Shawn foresees that intercepting one-time passwords will continue. Also, “we’ll likely see an increase in 2024 and 2025, and customers need the right tools before implementing self-service,” warns Shawn. “They also need tech that can flag high-risk calls,” he continues.The growth is due to fraudsters’ ability to adapt very quickly. Shawn says, “We’re also too reactive to something happening before controls are implemented. You need to take proactive measures.” It takes good technology moving at the same speed to detect fraud.
8. How can we trust voice authentication in today’s market?
   “Voice can be a trusted authentication solution with today’s technology, but the technology has to be able to identify liveness so that the voice can be trusted,” says Shawn. Pindrop’s new product Pulse was tested with 12 current Pindrop customers, and all returned saying it worked in identifying deepfake attacks in call centers, including zero-day attacks. Technology has to be multifactor to tell you if it’s a live or not-live voice.
9. How does Pindrop detect familial fraud situations?
   Familial fraud generally refers to deceptive practices carried out by family members against one another, often for financial gain. One example could be inheritance fraud, where a family member forges a will, manipulates an elderly relative to change their will, or hides assets to receive a larger share. It could also be investment fraud, where a family member might convince relatives to invest in fraudulent schemes or take money intended for legitimate investments.“This one can take a lot of work to catch,” says Shawn. However, the power of utilizing the Passport and Protect products together is very helpful.  Shawn explains that if the organization has an enrolled Passport profile for the caller and the Protect risk score, this information can be used to determine if a voice is a mismatch to the profile.  Additionally, as organizations provide negative feedback into the Protect risk engines, a negative voice profile will be created and compared to all subsequent calls. Familial fraud happens frequently in the insurance sector, but also happens within many other verticals as well, so it’s essential to have these technologies in place to prevent it from happening in the future, according to Shawn.
10. What should financial institutions be concerned with in the future?
    Shawn says fraud detection and multi-factor authentication are mission-critical in financial sectors. There’s a lot of cross-pollination, especially in this sector. People don’t bank with just one bank. When you’re exposed, you’re exposed across many institutions, no matter the size of the organization.

Final thoughts on rising trends in fraud and implementing security in 2024

As technology grows, fraudsters may strive to make their schemes as believable as possible. Innovative companies are now implementing proactive technologies to better catch these fraudsters. Early adopters are getting ahead of the curve. Being proactive reduces the likelihood that you have to take the losses in the future. For most customers, fraud can go from a nuisance to a malicious problem.

Request a demo to see how Pindrop technology can help you protect your brand from fraud and build better customer relationships in the future.