CSI: Atlanta? No, It's Phone Fingerprinting

A criminal dials into your company's call center, looking to steal data on one of your customers. He's pretending to be the customer, fishing to get a password, a mother's maiden name, or anything that might help him set up a fraudulent account.

About 18 seconds into the conversation, your company has identified the call as fraudulent. You know where the fraudster's calling from, what names and numbers he has used before, and even his calling patterns. The transaction is blocked, and the caller is identified and blacklisted.

Is this an episode of CSI? No, it's new technology from an emerging vendor called Pindrop Security, which received an $11 million round of venture funding from Andreessen Horowitz and other firms on Wednesday.

Pindrop has developed a patent-pending technology called Phone Fingerprint, which enables companies to identify fraudulent callers through forensic analysis of their calls. Phone Fingerprint analyzes information such as the phone number and device used, voice recognition, and even background noise to uniquely identify callers within 15 seconds. It then takes about three seconds to flag the company and terminate the transaction.

"A lot of criminals and fraudsters are finding that socially engineering an inexperienced employee or call center representative is a very effective attack vector," says Vijay Balasubramaniyan," co-founder and CEO of Pindrop. "Depending on human error to break into an account is a lot easier than most forms of online attack."

About 30 percent of all financial fraud begins with a phone call, Pindrop says, and some financial firms say that as much as 60 percent of the fraud they see takes place over the phone. About one in every 3,000 calls to a customer call center is a fraudster, Pindrop estimates.

"That doesn't sound like a lot, but what it means is that the average call center rep is probably not very experienced in recognizing a fraudster," says Matt Anthony, vice president of marketing at Pindrop. "Our goal is to recognize that fraudster and let [the company] do something about it before the transaction is completed."

Phone Fingerprint analyzes phone call audio signals to identify the caller's location and calling device type to create a unique fingerprint, which can be used to match the caller to other calls they've made, regardless of attempts to mask identity and calling activity.

After receiving a $1 million round of seed funding last year, Pindrop now has $11 million more to make its presence known in the enterprise arena.

"Financial institutions are on the front lines, facing a well-organized, well-funded growth industry of cyber criminals," said Arvind Purushotham, managing director at Citi Ventures, one of the investors that provided the funding. "Pindrop Security developed a truly unique technology, providing a legitimate solution to address two of the biggest problems financial institutions face today, detecting attackers and identifying legitimate callers."

Have a comment on this story? Please click "Add a Comment" below. If you'd like to contact Dark Reading's editors directly, send us a message.