A new wave of digital-first fraud is challenging insurers. The way to win the battle is through a one-two counterpunch of collaboration and advanced technology.
Learn more about 13th Annual RGA Fraud Conference and register to attend for free today.
Nichole Myers knows how to fight fraud. She also knows she cannot do it alone.
Myers is the chief underwriter at Ethos and a featured speaker at the 13th Annual RGA Fraud Conference, to be held online August 12-14. (More information and registration here.) In this interview Myers shared her view on the newest fraud threats facing the insurance industry, the tools necessary to tackle them, and the most effective approaches insurance executives can take to keep the bad guys at bay.
In today’s digital-first life insurance environment, fraud detection has become significantly more complex, and the stakes are at an all-time high. As insurers move toward streamlined online applications, fraudsters have adapted with equal speed, leveraging stolen personal information, bots, and now generative AI to exploit vulnerabilities across the policy lifecycle.
With application processes fully online, fraud attempts using stolen personal information to obtain life insurance policies have risen. A key concern is that these bad actors can bypass digital identity checks, establish fraudulent or misrepresented policies, and later attempt false claims or account takeovers.
As technology evolves, so do the tactics, which range from automated bots completing applications to subtle variations in personal details used to submit multiple applications. While the speed and convenience of online onboarding benefit legitimate customers, they also create new vulnerabilities that sophisticated fraudsters are increasingly exploiting.
Cross-system tracking and flagging is essential, and it is an area we are deeply committed to improving. In the life insurance space, critical signals are often buried across disparate systems – policy admin, billing, customer service. Each is typically reviewed in isolation by human teams. That fragmented view makes it difficult to detect subtle fraud patterns early.
This is where AI-based anomaly detection and behavioral pattern recognition come into play. These models are becoming increasingly sophisticated, but they are only as effective as the data and training we provide. It takes intentional effort to build cross-functional visibility and feed these systems with meaningful inputs.
For example, imagine a scenario where a customer submits a change of beneficiary or payment method shortly after policy issuance. That change is typically handled within an admin system, by operational staff, and might not raise any red flags on its own.
But an AI model trained on cross-system signals might flag this case because earlier customer service interactions showed urgency around rapid policy issuance without mention of any planned beneficiary changes. That discrepancy suggests the possibility of someone trying to secure coverage before drawing attention to potentially fraudulent intent.
In other cases, models can detect bot-like behavior during support interactions, looking for repetitive phrasing, unnatural timing, or suspiciously scripted language. Just as insurers are using AI to detect fraud, bad actors are using AI to commit fraud, automating interactions to slip through defenses unnoticed.
The key is aligning systems and data streams so that AI models can act as the connective tissue, linking what humans would otherwise see as unrelated activities and surfacing fraud before it does damage.
The biggest concerns are the "unknown unknowns." We have been very intentional about identifying gaps and asking targeted questions to understand what we are doing and why. But what should worry all of us the most are the risks tied to the questions we don’t yet know to ask.
We are trying to solve a puzzle with scattered pieces, some of which are visible, some that are still missing. It has been challenging to piece together a full picture, and we want to make sure we are not overlooking something critical simply because it hasn’t surfaced yet.
Ultimately, to stay ahead we have to stop thinking like defenders and start thinking like our adversaries.
Fraud is not a company-specific problem. Fraud is an industry-wide threat that crosses organizational boundaries, product lines, and systems, targeting the most vulnerable points. Allowing other companies to fight alone is not a competitive advantage. It creates opportunities for fraudsters to adapt and grow stronger.
Collaboration is not just smart; it is our strongest defense.
I don’t have all the answers. No one does. That is why we must work together. Fraudsters are getting smarter, faster, and more complex - and so must we. It is not enough to rely on legacy systems, siloed teams, or one-time audits. The most dangerous threats are the ones we don’t see coming, and they thrive in the gaps between our systems, our assumptions, and our questions.
We need to think differently, think across systems. think like adversaries, and – most importantly – we need to work together because no single company can win this fight alone.
The future of fraud prevention is collaborative, intelligent, and proactive. That’s where we need to be.
Nichole Myers’ presentation at the 13th annual RGA Fraud Conference is at 10 a.m. Tuesday, August 12. It is called 'The Devil Wears Data: Spotting Fraud in a World of Perfect Profiles.' Learn more about the conference and register to attend for free today.
