Canadian Underwriter
News

Here’s a way small brokerages can afford AI


June 4, 2018   by Jason Contant


Print this page

Budget may not be a big concern for larger companies looking to incorporate artificial intelligence (AI) into their operations, but is AI cost-prohibitive for small, local brokerages?

“I don’t think you need a massive budget to do some of this stuff,” said Nick Milinkovich, leader with the insurance analytics practice at McKinsey & Company (Canada).

.

Speaking at Insurance-Canada.ca’s first Technology in Action seminar last week, Milinkovich said smaller brokerages may want to consider finding a partner to help them implement AI.

“There are lots of insurtechs that are focused on this problem. There’s a bunch in Toronto, actually, that are focused on exactly the challenge you put forward,” Milinkovich said in response to an audience question.

“Lots of tools are emerging to help those folks be more productive and successful in what they do,” he added. “It’s less about needing a big budget. It’s more, ‘Why do I need to participate? What is the particular thing I need to solve to keep [myself] competitive and successful and then finding the right partner because they can’t build it themselves.”

Four trends are shaping the adoption of AI, or advanced analytics, for both brokers and carriers:

  • Explosion of data: different forms, tools and types of data (including structured and unstructured, voice and video)
  • Connected devices: sensors throughout a home that detect which rooms people are in, and how they spend their time and electricity usage, for example
  • Ecosystems and open source of data: ecosystems are emerging (more so in Asia) that allow sharing of data between telecommunication providers, insurance companies, and Amazon, as an example. “If you think of your ability to share data across business units, that’s already tough enough,” Milinkovich said. “To think about how would I share data in a structured way with a totally different industry, that’s pretty scary.”
  • Advances in cognitive algorithms: operating in real-time and adapting to the nuances, biases and algorithms.

“If you put all four of those things together, that’s a lot of change that going to come and, for a lot of carriers that are trying to adapt, it’s a daunting starting point,” Milinkovich said.