Pindrop unveils latest call center anti-fraud solution

September 28, 2017 - 

Pindrop, a pioneer in voice-fraud prevention and authentication, today has launched the newest version of its flagship offering.

The firm today released Pindrop Protect 4.0, a multi-factor anti-fraud solution that allows call center operators to assess the fraud risk of every incoming or outgoing call. The solution ensures that every facet of a call, including audio, voice, and metadata, is scored based upon anomaly detection and matching. The solution is noted to provide deep call intelligence for every call in real-time, along with fraud call detection for first time callers.

The solution leverages Pindrop’s newest version of its patented “Phoneprinting” algorithm that can analyze a phone call in 30 seconds to identify malicious behavior and verify a legitimate caller.

Phoneprinting is at the center of Pindrop’s Call Center Anti-Fraud solutions, used by a wide array of Fortune 500 companies to detect fraud and verify customers. This technology utilizes shared customer blacklists, along with the world’s largest audio database of fraudulent calls and artificial intelligence (AI) machine learning to provide a highly accurate identification of fraudsters.

The previous version of the technology analyzed 147 unique audio characteristics. This next generation of Phoneprinting technology included in Protect 4.0 now analyzes over 1,300 audio features of every phone call to create a “high definition telephony print” while detecting anomalies that indicate fraud.

According to Matt Garland, Pindrop’s VP Research, in-depth granularity from the analysis of a greater amount of features creates high-resolution telephony profiles that can deliver up to a 20 percent accuracy improvement in catching the sophisticated, repeat fraudsters that impact a customer’s bottom line. The firm notes that a clearer picture of these dangerous fraudsters enables Pindrop to catch an additional US$50 million in fraud losses for Pindrop’s existing customer base.

Phoneprinting works by determining a caller’s true geographic location, device type, and carrier regardless of spoofing. Phoneprinting also brings responsive recognition for the most sophisticated of fraudsters and highly accurate anomaly detection even for a fraudster’s initial calls.

“We examine the unique characteristics that are in the phone calls that come into call centers,” Garland said in an exclusive interview with BiometricUpdate.com. “We derive in our product tiny breaks in the audio that indicate VoIP packet lost, along with noise profiles that uniquely identify callers and their device. We have utilized our technology to analyze over 400 million calls and our Ph.D. research staff, with this version of Protect, have found a method to deliver about 10 times the clarity and insight into these calls. This is the ‘4K’ equivalent of our Phoneprinting technology.”

Pindrop’s own research shows that call centers in the U.S. receive a fraudulent call every second and that call center fraud has been rising more than 200 percent over the last few years. Fraud losses, as a result, totalled US$14 billion last year. Companies are also spending an additional US$8 billion annually on authenticating customers. The firm’s latest solution allows call centers to reduce reliance on knowledge-based authentication questions. This approach allows Pindrop to continue to detect fraudsters that attempt to hide their identities.

“We know on average that for every call coming into a call center there is a 58 cent loss due to fraud,” said Garland. “Collectively our customers, which are 70 percent comprised of the Fortune 500, will collectively save $50 million in fraud losses from our solution upgrade. So for enterprises with large call centers they will experience millions in savings, from reduced fraud loss and operations costs.”

The company notes that reductions in its client’s call center operating costs will accrue from less human and financial resources being dedicated to fraud prevention and remediation due to improved fraud detection through the service.

Further, utilizing multi-factor technology, the solution will also leverage more traditional biometric technologies. Call centers using Protect 4.0 will leverage Pindrop’s voice biometric technology that passively analyzes speech to crossmatch a caller’s voice against a list of known fraudsters. Using voice printing, Pindrop can identify repeat fraudsters no matter if the voice changed, altered or modulated and add them to a fraudster blacklist. Pindrop Protect use this fraudster blacklist to quickly spot fraudsters on repeat attacks.

In a statement concerning the new solution’s release, Vijay Balasubramaniyan, CEO and co-founder of Pindrop, said: “As costs and losses continue to rise, the C-Suite is waking up to the fact that phone security is becoming the nexus of fraud across their organizations. Protect 4.0 combines six years of data, knowledge and insights to deliver a second generation hi-def phoneprint that will make it very difficult for bad actors to continue committing a faceless crime.”

Leave a Comment

comments

About Rawlson King

Rawlson O’Neil King is a contributing editor at BiometricUpdate.com and is an experienced communications professional, management consultant, trade journalist and author who recently published a book about control and electronic networks and who has written numerous articles in trade publications and academic journals about smart home and building technologies. Follow him @rawlsonking2.