3 ways AI could transform your insurance policy

The technology that saturates today's world — smart-home devices, drone images, fitness trackers, social media posts and telematics programs that monitor your driving habits — can help insurers piece together a detailed picture of your behavior.

Your permission isn't always required. Many facts about your house, car and neighborhood are public records. Data brokers also gather and sell details about your activity, like which stores you visit, what you click online and the whereabouts of your mobile phone.

For a human, all that data is too much to process. But the ability of artificial intelligence to interpret data could upend the process of buying an insurance policy and filing a claim. As insurers face questions about fairness and privacy, some people may find it's harder to get coverage. Others will benefit from cheaper rates, quicker applications and easier claims.

Faster insurance applications

Customers could see a shortened application process as insurance companies embrace AI.

Insurers may drastically cut the number of questions they ask in a home insurance application, says Peter Flynn, head of personal lines for the Americas at insurance consulting firm Xceedance.

"In the future, they might only ask five questions," Flynn says. "But they might gather 5,000 additional data points, and they might interpret those 5,000 things in addition to the five answers they get from the applicant."

Chicago-based Kin Insurance, for example, collects thousands of data points and "prefills" home insurance applications with property details like square footage, foundation type and number of bathrooms.

A similar shift is happening in life insurance underwriting, which traditionally requires a medical exam plus a health and lifestyle questionnaire. As AI models improve, more carriers offer accelerated underwriting — quickly issuing policies to low-risk customers based on digital medical records and other data, while flagging higher-risk applicants for conventional underwriting.

"You can put a little bit of information and they can return a rate that's not based on somebody coming to your house and taking blood," says David Embry, CEO of online insurance broker Mylo.

To get the most accurate rate, make sure your records are correct and up to date before starting a life insurance application. You might also want to have supporting documents — like a summary from your doctor about any medical conditions — ready to go.

More personalized insurance rates

Low-risk customers stand to save money as insurers use data to create increasingly personalized profiles of their users.

The auto insurance industry is leading the charge with telematics programs that monitor things like your speed, braking patterns and mileage, enabling insurers to base pricing on driver behavior.

"In an AI-enabled or machine-learning-enabled environment, they can take that to an infinite degree and gather and collect as much data as available and interpret it in real time," making predictions based on an individual's habits, Flynn says.

While low-risk buyers reap the benefits of a constantly fine-tuned prediction model, a 2020 report by the Organization for Economic Cooperation and Development warns of the potential downside of this approach. Slicing and dicing customers into smaller risk pools could effectively price some applicants out of insurance, the OECD report says.

For drivers, the smart approach is to compare car insurance quotes from several companies. Insurers don't all use the same sources of data, and they weigh each factor differently.

Simpler claims, and maybe fewer of them

Filing an insurance claim can be a stressful experience. Insurers' use of AI could make the process smoother for customers and get them a decision — and their payout — much more quickly.

AI can help insurers identify the most urgent claims, reconstruct accident scenes, analyze medical records and flag cases for signs of fraud, according to a 2023 report by research firm Everest Group and professional services company Ernst & Young. Making claims more efficient is a priority for more than half of the property and casualty insurers surveyed, the report says.

New York-based insurer Lemonade says AI-based insurance fraud detection allows about 40% of its claims to be resolved within moments.

AI could even help prevent losses before the need for a claim arises — known as a "predict and prevent" model instead of the current "detect and repair" approach. For example, data relayed by smart-home devices could automatically trigger intervention if, say, a sensor catches early warning signs of a leak or a frozen pipe.

AI can also deliver feedback to drivers, helping them adjust their behavior. Programs like Allstate's Drivewise reward those who avoid risky habits like speeding, hard braking or using a phone while driving.

But as the insurance industry integrates AI, there are concerns about cybersecurity, privacy and the potential for AI models to discriminate based on characteristics like race or gender.

The National Association of Insurance Commissioners issued guidelines in December 2023, encouraging insurers to correct errors in AI models and avoid bias. But each state creates its own rules, and regulation remains in its infancy stages.

Oversight will evolve, Flynn says. "But I'll bet you the technology evolves faster than the regulation."