Carpe Data urges industry-wide collaboration to stem AI-driven insurance fraud
An American data and analytics firm is calling for industry-wide collaboration to find solutions for artificial intelligence-driven insurance fraud, which is becoming increasingly complex and widespread.
“It’s always been this way. People have always wanted to try to commit fraud; they would submit either fake documents or altered receipts. But with AI, it not only allowed more people to be able to do it, it takes very limited knowledge to use some of these tools, but also the real thing is the volume of it,” Tom Rasmussen, vice president of product at Carpe Data, said.
Although AI advancements have opened the door to countless opportunities for the insurance industry, the same can be said for bad actors, who now suddenly have a way to exploit vulnerabilities more easily and at a scale “much more so than they ever could in the past.”
“My main worry is that things are moving so fast. How do we keep up in such a highly regulated industry? It’s about open communication on both sides to make sure that we’re understanding the problem and finding solutions for it,” Rasmussen said.
New trends in AI fraud
Rasmussen noted a host of new ways bad actors are using AI to conduct insurance fraud, including making it appear there is more damage than actually exists, generating damage that isn’t there, or generating false receipts or invoices.
“Those are standard, but a new thing that we’re seeing and talking to carriers about is using AI to spoof the carrier’s own website. What they’re doing is using coding agents and AI to generate their own websites that look exactly like the real thing, dupe unsuspecting customers into purchasing policies and submitting payments for premiums, all that stuff,” he said.
He noted that it is not necessarily a new occurrence, as “there have always been street brokers and this kind of scheme of people selling policies, acting like they’re an agent for that broker.”
The difference this time, however, is that AI now enables them to do this online at the click of a button.
“Not only can they make those sites, but they can then also use SEO techniques and AI to generate the traffic and redirect it there using keywords and those sorts of things. It just compounds, and makes it that much more at volume and at scale. AI allowed people with no coding skills or very limited Photoshop skills to suddenly have the tools at their fingertips. Therefore, you need more solutions to help solve them,” Rasmussen said.
A call for collaboration
A bad actor will likely not stop at just one provider but will attempt to dupe others, Rasmussen suggested. This, along with the scale and speed of AI-driven fraud, necessitates collaboration and communication to weed out the problem.
“The hardest part from a carrier standpoint is that we work in a very highly regulated industry. Within each carrier, there are generally layers of management and procurement and different safety systems, and the problem is that all those processes take time,” Rasmussen noted.
“Meanwhile, the bad guys don’t know the rules. They have no safeguards slowing them down. They are just full steam ahead, and so they can be much more nimble; they can move much more quickly.”
For him, the key is communication between vendors and partners at insurance carriers to share information pertaining to fraud trends, flagged individuals or cases, and workable solutions.
“If I’m able to hear about issues they’re seeing and they’re able to surface those to me, then I can hopefully implement some sort of safeguard on the other end to help protect them and take more of a defensive posture or even a proactive kind of offensive posture, depending on what the problem is,” Rasmussen said.
This will help the insurance industry “move that much quicker and hopefully keep up with the other side that’s, unfortunately, able to move at light speed right now.”
“Unfortunately, in the US, we don’t do a good job of prosecuting insurance fraud. I know every once in a while, there’s an article here or there, but that pales in comparison to the amount of fraud that never gets prosecuted,” Rasmussen said.
“They literally just pick up where they left off and move to a different carrier or a different state or a different place to try again. And so, the more we can share that data, the better.”
AI-powered solutions
Putting action to those words, Carpe Data has already released some solutions to help fight AI-generated insurance fraud. It currently offers fully integrated software with Verisk’s ClaimSearch platform that flags potential fraud and sends alerts out.
Rasmussen also hinted at a suite of new tools in development to be released at a future date that aim to provide “nice, seamless integration with no IT tech lift.”
“Specifically for AI, we’re trying to leverage it on our end to get some new products out in the marketplace. And that’s a really powerful message to make sure that we can stop the repeat offenders, because they certainly don’t stop with just one person,” he said.
© Entire contents copyright 2026 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 Carpe Data urges industry-wide collaboration to stem AI-driven insurance fraud appeared first on Insurance News | InsuranceNewsNet.
