New technology uses AI to help insurers detect fraudulent activity

A new AI-powered solution has emerged to help insurers detect fraud by flagging unusual user activity directly on an insurance company’s website.

Glassbox is a digital experience intelligence platform that allows companies to monitor user experience on their respective websites. Its standout capability, the Glassbox Insights Assistant, is an agentic AI that helps financial companies process and act on engagement data.

Joe Crawford, director of professional services, explained that the AI quickly sifts through the large amount of website data, understands which behaviors are normal and flags abnormal activity such as unauthorized logins, suspicious changes to policies and more.

“We deal with a lot of large insurance companies and AI is really helping us to sift through the large amount of data that we have, that we’re using to monitor the user’s experience,” Crawford told InsuranceNewsNet.

He noted that on a typical website, hundreds of thousands of people visit or use those webpages every day. It is “impossible” for a team of trained staff to have to constantly monitor all of those sessions and identify the small details that make up fraudulent attempts, he said.

“There’s a lot of good people trying to do good things out there, but then there are some bad actors out there that are really trying to take advantage of certain situations. AI is helping us to distill all of that data and understand the anomalies in the data, to sift the wheat from the chaff or separate the sheep from the goats,” Crawford said.

Increasingly complex fraud

In today’s landscape, insurance fraud is becoming increasingly complex and evolving as new digital tools emerge, Crawford noted.

“There are a lot of very savvy fraudsters out there. They’re doing things that we wouldn’t typically expect… We’ve seen where people are doing things like making changes to the actual website on the fly, taking screenshots and sharing it with support personnel within an insurance company,” he said.

In his view, this is why using technology like AI-powered digital experience monitoring tools is non-negotiable when it comes to identifying bad actors and preventing related loss.

“People who don’t have these digital experience monitoring tools or are not made aware of what’s happening, they can be taken advantage of and it costs insurance companies money. But now that they have these reports and we’re using AI to help distill that information to bring all of that data into a summarized view, they’re able to see and alert the authorities, save a lot of money and mitigate a lot of risk,” Crawford said.

Analyzing user activity

Glassbox’s AI works as an embedded solution directly on a client’s website. It continuously monitors everything a user is doing on the website, including clicks, mouse hovers, interactions, form submissions, etc. That activity is also recorded as a replay that can be viewed again.

All of the captured website data is sent automatically to a dedicated Glassbox server for the client, where internal users such as the company’s support, fraud or security teams have full access to replays of user sessions and other AI-generated insights. The Insights Assistant also summarizes all captured sessions and generates reports based on certain patterns of behavior within each session.

“We program the AI to [recognize] what’s considered normal and what’s considered not. It looks at patterns of sessions that are considered atypical within all the sessions that it actually collects, and we configure it to send anomaly alerts and things like that to the customer’s personnel teams,” Crawford said.

He added that it’s particularly useful for compliance or legal teams, who can quickly check where users are having issues or see where fraudulent activity is taking place, with the replay to actually prove it.

“Some people are able to use this even in the court of law to show that we can actually see the discrepancy of what the user is saying to what they actually did,” Crawford said.

All about user experience

Despite Glassbox’s anti-fraud applications, the platform was primarily created to help insurers and financial businesses understand their customers’ digital experience and improve that over time.

“Even outside of fraud, it’s important to really understand the customer’s experience…especially in insurance where they’re setting up claims and things like that, just because there’s usually an emotional attachment when people are submitting claims — either there’s a health issue or they had an issue with their cars and things like that,” Crawford explained.

In today’s world, he said, customers are looking for personalized, frictionless digital experiences and they have multiple options to choose from. Customer experience is what differentiates one company from a competitor and makes them stand out, he added.

“So, the fraud is one part of it, but making sure that the insurance companies understand the customer’s experience and address that in the correct way is super important as well,” Crawford said.

Glassbox is a digital experience intelligence platform for insurers and financial institutions. It was founded in 2010 and is based out of London, United Kingdom, with offices in New York and Israel.

© Entire contents copyright 2025 by InsuranceNewsNet.com Inc. All rights reserved. No part of this article may be reprinted without the expressed written consent from InsuranceNewsNet.com.

The post New technology uses AI to help insurers detect fraudulent activity appeared first on Insurance News | InsuranceNewsNet.