Credit: Urupong/Adobe Stock
For years, it was surprisingly easy for inaccuracies to slip through on insurance applications, often going unnoticed until a claim was filed, at which point the consequences could fall squarely on the insured.
In the landlord insurance space, that might mean saying a roof was replaced more recently than it was, underreporting the number of units in a building, or leaving out a commercial tenant. Sometimes it’s intentional. Often, it’s not. Many property owners just don’t know the exact answers, and insurers haven’t historically asked too many follow-up questions.
But the consequences are real in this $60 billion market, where roughly 20% of applications contain misrepresentations about roof age alone. Across the broader insurance industry, non-health insurance fraud costs more than $40 billion a year, raising annual premiums by $400 to $700 for the average family. It adds up quickly, and honest policyholders end up footing the bill.
That’s where AI is starting to make a difference.
At Honeycomb, we use a mix of aerial imagery, street-level photos, and public records to automatically check what’s on an insurance application. Was the roof really replaced last year? Does the building have 12 units or 20? We’re able to answer those questions in seconds, without sending someone to the property. That cuts down on guesswork required by the insured during the submission process and gives us a more accurate picture of risk from the start.
Interestingly, the original goal of this technology wasn’t fraud prevention. It was about making it easier to quote properties and to streamline the underwriting process. But better information at the outset also helps prevent inaccurate pricing and reduces the need for costly manual inspections or dispute resolution based on unintentional accuracies down the line. What makes AI so effective at detecting inaccuracies and fraud is its ability to spot patterns and, more importantly, anomalies. By analyzing millions of data points across properties, locations, and claims histories, AI systems learn what “normal” looks like.
They can flag outliers quickly, whether it’s an unusual roof shape that doesn’t match the reported condition, square footage that seems off compared to similar buildings, or a cluster of claims that follow an unusual pattern. These are discrepancies that humans could miss, especially at scale, but they jump out to a machine trained to look for them.
And critically, AI does this in real time, without adding friction to the customer experience. It’s not about treating everyone with suspicion. It’s about using better tools to verify, so we end up with a more accurate policy that allows all parties to win.
The impact has been tangible: our loss ratio, the amount paid out in claims compared to what we collect in premiums, is significantly below the industry average. That tells us we’re catching discrepancies early, underwriting more accurately, and providing “right-priced” quotes and consistent pricing and experience in a market where insurance prices have gone up significantly.
The broader industry is moving in the same direction. Nearly 60% of insurers are now using AI to flag suspicious claims, and the fraud detection market is expected to grow to $17 billion by 2028. As the cost of doing business continues to rise, due to climate events, inflation, and other factors, accuracy and fairness are becoming more important than ever.
Most applicants aren’t bad actors, just don’t have all the information, and the process hasn’t made it easy to be precise. Legacy insurers have never figured out how to guarantee the integrity of self-reported application data, effectively training customers that they can put whatever they want, and nobody’s really going to notice.
With AI, we can set a higher baseline for accuracy and trust. And that’s good for everyone. At the end of the day, it’s the honest customers who end up footing the bill for all the inefficiencies in the system, and misrepresented applications are a big part of that. By fixing this, we’re making the system fairer and making sure honest customers come out ahead.
Itai Ben-Zaken is the Co-founder and CEO of Honeycomb, a tech-forward digital insurer redefining the multi-family property insurance landscape by providing instant, customized, affordable coverage through its end-to-end digital platform.
© Arc, All Rights Reserved. Request academic re-use from www.copyright.com. All other uses, submit a request to TMSalesOperations@arc-network.com. For more information visit Asset & Logo Licensing.