How AI Will Reshape the Insurance Industry

How Insurers Are Using AI in Their Operations

Insurers are using or plan to use AI up and down the value chain in their operations. It can help companies optimize their marketing dollars by predicting who will respond to which types of outreach.

It can also be used as a component to generate a property risk score, even though regulators frown on it being used as the sole or primary factor at arriving at such a score. “What you never want to use AI for is something that marginalizes or creates harm to any one group,” says Russell Page, CTO of Hagerty, an automotive lifestyle company and a leading specialty insurance provider for the global collector vehicle market.

However, he says, in the future, AI could be used to create a more personalized risk assessment for customers without using broad-based proxies for risk, such as someone’s credit score. Hagerty is exploring using AI around rating, claims and fraud detection, among other areas, Page says.

Indeed, AI can be used to process images and determine the cost of fixing damage, McElhaney says. It also can be used to detect fraudulent claims more effectively than human claims -adjusters can, according to McElhaney. Insurance claims experts are primarily going to rely on their experience to determine whether a claim is fraudulent. However, an AI-based tool can be trained on reams of data to determine common patterns of fraud.

“AI has the ability to discern patterns in ways and in data sets where humans simply cannot, or they simply just don’t have the capacity to look at ginormous data sets and tease out various patterns,” McElhaney says. “That’s what AI can do very successfully.”

The Role of AI in Reducing Friction

AI can deliver numerous benefits to insurers, and many of them boil down to reducing friction in decision-making, both internally and for customers.

“You reduce the number of questions that you have to ask,” Page says. “You shorten the time to service a claim where you’re keeping the promise, and you find ways to take costs out of it so that ultimately you can lower premiums.”

The goal, Page says, is to create value for customers and do so in a way that “maintains privacy and compliance across the board.”

EXPLORE: The role of predictive analytics, edge computing and zero trust in 2023.

AI-based tools can help insurers increase response rates to marketing campaigns by identifying the attributes of those who responded and then targeting similar customers moving forward, McElhaney says.

Such tools can also be trained on insurers’ claims data to determine whether a claim is highly complex or severe or whether it is of low complexity and severity. The latter kinds of claims, which are much less likely to be fraudulent, could be paid out automatically, he says.

Relying on AI to Lay the Foundation for Modernization

How easily insurers can leverage AI will vary depending on their existing IT infrastructure and the skills of their workforce, as well as the cultures they have developed internally, according to Page and Hagerty.

Some insurers are using “very legacy systems and will have a hard time modernizing and creating that type of modular and scalable approach to adding capabilities, and others have started a modernization journey” and will more easily benefit from AI, Page says.

Claims makes up roughly 70 percent of the costs of any carrier, Page says, and “if you can reduce that cost, you’re going to be able to give that back in terms of rates or increase the profitability of the organization.”

DISCOVER: Why banks should do more with their rich troves of customer data.

Many insurers are “fairly sophisticated and are fairly proud about their technical capabilities,” according to McElhaney, and have invested in data scientists. At the same time, there are many technology companies that have built specific AI-based tools, such as those for fraud detection, and want to sell directly to insurers.

“All carriers will go through this evaluation exercise of, ‘Do we have capacity to build the capability? And even if we do, should we? Because we can buy something that’s better and we can allocate our resources otherwise,” McElhaney says.