Canadian Underwriter
News

At least two cool ways that AI can help insurance companies


February 28, 2018   by Jason Contant


Print this page Share

Machine learning could be used to underwrite faster, reduce fraud and assess vehicle damages more accurately, Lawrence Wong, Munich Re Canada’s director of application development, said Tuesday.

Speaking at the 16th annual Insurance-Canada.ca Technology Conference in Toronto, Wong said that artificial intelligence (AI) is here to stay and will augment the way people do the business of insurance, whether insurers leverage it or not.

“The power of machine learning is that you don’t have to explicitly program it,” Wong said during the session Case Studies in AI. “As you show [a self-learning machine] more and more pictures of bumper damage, it will, with experience, become more accurate at defining what’s vehicle damage versus what’s quirky vehicle design from the car manufacturer.

“Notice I said the word experience,” Wong added. “That’s exactly how a human loss adjuster would learn about vehicle damage. AI is about augmenting people.”

In a separate session, Charles Dugas, director of insurance solutions at Element AI, said vehicles are already currently highly connected with different sensors.

“So what you could try to do is assess the damage in real-time, leveraging sensor data,” Dugas said. “Obviously, that would require working with manufacturers themselves in order to understand how and where the sensors are placed and what kind of data can be related to shifts in the sensor.”

Through the sensor, an adjuster can assess internal damage to the vehicle, such as the engine. If the damage is sufficient to say the car cannot be driven, customers could be advised in real-time that it is a total loss.

“If that’s the case, you can in real-time notify your customer that a tow truck is coming, and that they shouldn’t drive the car,” Dugas said. “They will be taken to a certain garage and a rental car would be waiting for them at that point. In a case where this is too close to call, you can revert to pictures. And if that’s too close to call, again, you may need human appraisal.”

Assessing vehicle damage is not the only advantage of AI. Munich Re uses an application called Internet Research & Intelligence System (IRIS).

“You feed it a search [term],”Wong explained. “Then it goes out to the Internet, pulls down social media, the news, fake news, documents and articles, and presents all the results in a word cloud.”

For example, in one search, terms like “breach, data leakage, North Korea, hackers [and] CIO resigns” appeared, Wong told delegates. “This tells me one of two things: 1) it’s a red flag, we don’t want to insure that company or 2) that’s an opportunity, we should offer them cyber insurance because judging from their headlines, they could definitely take advantage of it.”