Pindrop® Pulse Helps Leading U.S. Insurer Protect Against AI Threats

As a Fortune 100 company offering a full range of insurance and financial services across the U.S., the company’s leadership saw deepfakes and AI-driven voice fraud as imminent threats to their business and members. They turned to Pindrop to help future-proof their fraud defenses before these threats increased.

Our Pindrop^® Pulse solution helped the insurer to better protect against future AI threats. At the same time, Pindrop^® Protect multifactor fraud detection solution provided a strong framework to help protect their members from fraud.

About the insurer

Formed almost 100 years ago, the insurer is now one of the top 10 mutual insurance companies in the United States and one of the largest insurance companies in the Fortune 500 list. It offers a wide range of insurance products such as car, motorcycle, homeowners, pet, farm, life, and commercial insurance, as well as financial services products like annuities, mutual funds, retirement plans, and specialty health services.

The insurer has led the industry by boasting several “industry-first” initiatives, including endorsing Medicare before being required by law, establishing 24-hour claims reporting service, and simplifying the sales and underwriting process for variable universal life insurance.

As a member-first and partner-oriented organization, the insurer was aware that new AI technologies and deepfakes could accelerate fraud, and that it was necessary to develop controls to safeguard member assets and build credibility with partners. As a long-time customer of the Pindrop^® platform multifactor fraud detection and authentication solutions, the insurer sought Pindrop’s help in making their business future-ready for AI threats.

Challenges

The insurer is a leading provider of 457(b) retirement solutions and manages employer-sponsored deferred compensation plans for 15 U.S. states. Their government business (retirement, annuity, life sectors) operates through multi-year, RFP-driven contracts that require rigorous due diligence and compliance with regulatory standards.

Their customers, such as the Kentucky Deferred Compensation plan with $4B in assets, were interested in how the insurer is protecting their members from deepfakes. Other plan sponsors and senior leaders across business lines increasingly asked about deepfake risks and the insurer’s strategic response.

This due diligence process sparked the insurer’s pursuit of a deepfake detection solution.

Goals

The insurer had three goals as they sought a new solution:

With these goals in mind, the insurer wanted to strategically adopt deepfake detection as part of the organization’s larger security framework and demonstrate to their partners, members, and customers that they were prepared to protect their interests from deepfake threats.

Before Pindrop^® solutions

The insurer was already using Pindrop^® Protect to detect fraud in the contact center, and were confident that the solution would help them catch deepfake fraud in the back-end. But the team was concerned about the growing threat of AI and the speed at which the AI technology was advancing. It was becoming increasingly easy for fraudsters to create believable synthetic voices and to have those voices say anything they wanted, thus creating an environment of distrust.

The insurer decided to get ahead of this threat by investing in a dedicated deepfake detection solution that would specifically aid their security teams by highlighting the presence of AI fraud in their contact center.

This concern of the company and its partners is well-founded. Text-to-Speech (TTS) models, the fundamental building blocks of deepfakes, have exploded in both number and sophistication. The number of TTS models listed on Huggingface grew from 300 in 2023 to 3,300 today—11x (1,100%) increase in just two years. Massive technological advancements like Nvidia’s Riva Magpie zero-shot voice cloning allow synthetic speech tools to generate a desired voice using just a few seconds of reference audio, making it easier to perpetrate fraud at a large scale.

Pindrop data from the 2025 Voice Intelligence and Security Report found that deepfake attacks on contact centers have grown 1,337% since 2023, and synthetic voice fraud in banks and insurance companies has spiked by 149% and 475%, respectively, in 2024.

The insurer recognized that it was time to fight fraudsters with their own tools—AI-based deepfake detection.

The solution

The insurer participated in the Pindrop^® Pulse beta program and rigorously stress-tested the solution in multiple scenarios involving deepfakes. The team acted as fraudsters and attacked the system with different techniques to test whether the solution would hold up. Once they were satisfied that Pulse was able to accurately detect deepfake attacks, the insurer chose to roll out Pulse into their contact center.

The insurer’s team also valued that Pindrop^® Pulse was “plug-and-play.” The team was able to incorporate the solution into the existing fraud detection framework and start operating as desired. Pindrop customer success team worked collaboratively with the insurer’s team to tune the solution to their needs and onboard it within the current policy environment.

Why did the insurer choose Pindrop^® Pulse?

The results

The insurer deployed the Pindrop^® Pulse solution in their contact center in September, 2024 and saw a significant impact.

Accurate deepfake detection

Pulse not only helped detect deepfake fraud at the front end, it also helped stop suspicious callers from enrolling and exploiting the authentication process, without increasing the administrative burden on the organization. The insurer is now able to identify deepfake attempts in real time and immediately put up guardrails to help prevent bad actors from perpetrating fraud. This has helped prevent financial losses, protect member assets, and clearly identify attempted deepfake attacks, which the insurer was unable to do in the past.

Deepfake detection from multiple attack vectors

Pindrop^® Pulse AI-based liveness detection provided the ability to detect various types of deepfake attacks, including:

This helped the organization to detect multiple attack vectors and gain a deeper understanding of fraud methodologies.

What comes next?

Video deepfakes, particularly those in virtual meetings, are the next frontier of fraud. As attackers evolve, their targets will move beyond the contact center to other areas where organizations are vulnerable, such as in executive impersonations, HR interviews, hiring and employee onboarding, and wealth management. Being keenly aware of these risks, the insurer is participating in the Pindrop beta program for Pindrop^® Pulse for meetings video deepfake detection solution.

About Pindrop^® Pulse

Pindrop^® Pulse is an industry-leading audio deepfake detection solution that, when combined with Pindrop’s multifactor platform, can detect up to 99% of synthetic voice fraud in the contact center. Pindrop Pulse technology seamlessly integrates with our fraud detection and authentication platforms, and is meticulously designed for real-time deepfake detection across all stages of a phone call. The award-winning Pindrop Pulse technology is built with proprietary datasets, 300+ Pindrop patents, and is backed by an exclusive Pindrop^Ⓡ Pulse Deepfake Warranty (terms and conditions apply).

Pindrop^® Pulse for meetings uses industry-leading deepfake detection technology that leverages multi-modal analysis to detect synthetic audio and video threats. Seamlessly integrating with major meeting platforms, Pindrop^® Pulse delivers continuous, real-time analysis for your most critical business interactions.