BI’s Article search uses Boolean search capabilities. If you are not familiar with these principles, here are some quick tips.

To search specifically for more than one word, put the search term in quotation marks. For example, “workers compensation”. This will limit your search to that combination of words.

To search for a combination of terms, use quotations and the & symbol. For example, “hurricane” & “loss”.

Login Register Subscribe

AI simplifies process for comp claims


SAN DIEGO — An overseer of workers compensation for the large grocery chain Wegmans Food Markets said AI, which typically stands for “artificial intelligence,” translates into “assisted intelligence” for his company, which saw litigation rates go from 40% to 22% among workers in high-risk distribution centers thanks to insurtech products put to work in recent years.

Using a program that mines employee emails, medical notes, claims handler notes, and phone calls, Wegmans knows that an “injured worker has expressed that they do not feel that they are recovering properly” and thus scores high on risk and needs intervention, Brian Trick, Wegmans’ Rochester, New York-based director of claims services, explained during a session Monday at Riskworld, the Risk & Insurance Management Society Inc.’s annual conference.

Mr. Trick displayed a screenshot of the app that assesses the injured worker’s mindset, engagement and employment status.

“Assisted intelligence for our claims team assists them to do their job,” he said. “It’s not going to replace them or do their job. It helps them to review volumes of medical records and has the ability to collect from different data sets and provide information.”

Workers can also gain improved access to treatment and better engage with the claims handler, Mr. Trick said.

“It really helps us get the best outcomes in lightning speed,” he said. “It doesn’t take us six days to authorize treatment. It can happen in real time.”

Tyler Kennedy, Boston-based lead developer for Gain Life Inc., which provides AI services in the comp industry, said no AI model is perfect, and that a human is needed to address program shortcomings. AI can misinterpret findings, for example, he said, using a grid featuring alternating close-up photos of a blueberry muffin and that of a chihuahua that an AI program couldn’t decipher to prove his point.