May 14, 2022

How AI is Changing the Paradigm of Insurance

An article in the May 2022 issue of National Underwriter by Daniel Turgel outlines “How artificial intelligence [AI] is moving the insurance industry in a bold, new direction.”

In the past, the expectation of an accident occurring at some point was pretty much the main reason motivating a business owner to buy insurance. Until the advent of AI, the only exception to this rule has been Mechanical Breakdown insurance (see our article in this issue), where loss prevention is the principal commodity. Now AI is set to change all that for most other insurance lines as well. Though losses will still occur, they should occur far less frequently, and the main advantage of having insurance may well be the AI that comes with it.


Human error is responsible for about 1.3 million automobile deaths worldwide each year: 38,000 of them in the US with an annual cost of $55 billion in medical and related expenses. As self-driving technology improves, it’s expected that nearly 90 percent of these deaths and costs can be eliminated. The other 10 percent? The Insurance Institute for Highway Safety says some remnants of human error will continue to operate and be responsible for over 90 percent of those fatalities, something it calls “final failure.”

But there are AI applications for “final failure,” too. For example, to reduce losses caused by truck drivers, commercial tech insurance companies collect data about the truck’s operation and the driver’s behavior that can reveal careless or even malignant behavior which deviates from the norm. In turn, drivers are coached into correct behavior that helps them avoid making careless mistakes such as “falling asleep at the wheel, backing up in a tight space with unseen obstructions, or spotting a car approaching at an unsafe speed and alerting truckers to the best way to avoid an accident.”


A water loss study by LexisNexis found that smart technology provided by the Internet of Things (IoT) can eliminate 90 percent of water damage losses by constantly monitoring facilities with water and fire prevention devices. What’s more, it’s expected that if there is a flare up somewhere, a robot will simply grab a fire extinguisher, walk up to any devices that are sparking and douse the flames.

Workers Compensation

The biggest problem with workers comp losses is not that safety measures and protocols have not been established. It’s that these measures and protocols are sometimes breached. Now, though, companies like Safesite, Pillar Technologies and Guardhat, using sensors and protective equipment, can address these compliance issues in real time with immediate feedback that prevents these breaches. Foresight Group, an insurance company, reports that these devices can reduce workplace incidents an average of 31 percent.

Insurance Fraud

Probably the most sophisticated artificial intelligence application for reducing insurance costs is pattern and anomaly recognition software. For example, one workers compensation third party provider, CodeBlue, has replaced the conventional First Notice of Loss process with what is called a “First Notice and Response (FNAR)” service. In this approach the claims specialist and the policyholder “collaborate” to “eliminate unnecessary demolition and reconstruction, limit unnecessary travel time and create efficiency, allowing adjusters to settle claims more quickly.” Most importantly, fraudulent behavior can be detected by capturing an image of the claims site before “unscrupulous contractors can get in and begin unnecessary repairs.”

Eventually, AI’s payoff should be far fewer accidents and much lower costs.

Article Courtesy of