What Can AI Do for Underwriters?

Artificial intelligence can now write books, create code, and generate art. But what can AI do for underwriters? We’ve already looked at how generative AI can boost claims readiness; now let’s explore how AI assistants can help underwriters streamline decision-making, eliminate tedious tasks, and unleash their full potential.

‍

Mine Years of Data

Insurers are accumulating mounds of data – this can either elevate them or bury them.

‍

There’s a lot of talk about how analytics can lead to better decisions. However, without a way to ensure data quality, combine data from multiple sources, and turn data into usable insights, you have nothing more than a bunch of numbers. Gartner says low-quality data can lead to poor decision-making, which costs organizations an average of $12.9 million.

‍

AI could be just the tool underwriters need to leverage data effectively. According to the Financial Times, Zurich is already exploring how to do this. Specifically, Zurich plans to use ChatGPT to extract data from six years of claims descriptions and other documents to identify loss causes and improve underwriting. It’s an exciting idea and the sort of novel application of generative AI that could revolutionize underwriting.

‍

Analyze Data and Assess Risk

In insurance, accuracy is everything. Underwriters need to check submission information to ensure every piece of information is correct. That can be a time-consuming and mind-numbing task. If any discrepancies slip through the cracks, the resulting underwriting may be incorrect, which can cut into an insurer’s profits.

‍

Underwriters also need to be eagle eyed when searching for signs of fraud. According to the Coalition Against Insurance Fraud, fraud occurs in around 10% of all property and casualty insurance losses. Some fraud occurs during claims; other times, it is present from the beginning. For example, many people lie about their address or the number of drivers to secure lower auto insurance rates. Unfortunately, this often slips through the cracks. Both insurance companies and their honest policyholders end up paying the price.

‍

Once again, AI can help. AI is simply better than humans at spotting discrepancies in vast amounts of data. It can flag issues for underwriters to deal with, which saves time, improves accuracy, and eliminates headaches. The only ones to lose out are the scammers.

‍

Bolster Communication and Education

Communication is important in the underwriting process. For instance, underwriters need to communicate with brokers about applications. Sometimes, there are problems with a submission, which means some back-and-forth conversation is necessary before the insurer can approve the submission.

‍

Insurers also need to educate brokers and policyholders about potential risks and loss prevention. Not only does this provide more value to policyholders – giving them another reason to be loyal – it also helps prevent claims, improving the loss ratio.

‍

Unfortunately, there are only so many hours in the day. Answering the same questions over and over can become a time drain.

‍

AI can help. According to Insurance Business, Coalition is using AI to reduce cyber risk by improving education. CoalitionAI Broker Copilot can answer questions about cybersecurity best practices, coverage, and more, whereas CoalitionAI Security Copilot can help with details on cybersecurity vulnerabilities, coverage contingencies, and ways to resolve cybersecurity issues.

‍

A Day in the Life of an AI-Empowered Underwriter

What will work be like for the AI-empowered underwriter? Here’s a typical day:

‍

You need to review a submission. It’s for a contractor and there’s a lot of information to go through. The broker says the contractor needs coverage ASAP, but you have an important meeting coming up. 

‍

No problem. Your AI-powered workflow audits the application to check for discrepancies, automatically comparing key datapoints with third-party data sources in seconds. It confirms the accuracy of most data but flags one possible issue. 

‍

Instead of reviewing every field, you can focus your attention on that one potential discrepancy. Upon further review, you confirm that it looks like there may be a mistake in the application: some of the numbers don’t match.

‍

You need to make sure you have the right information, but you also want to proceed carefully. This broker sends a lot of business your way – you don’t want to insult her. Normally, writing an email that strikes just the right tone would take you a while, but you use your AI assistant to write and send a friendly inquiry in mere seconds. You receive a response with the correct information and the updated data point is ingested into your platform without the need for data entry. 

‍

With the details out of the way, you have time to make a thoughtful decision. The contractor receives coverage, the broker is happy, and you still have time leftover to prepare for the important meeting.

‍

Do you want to see how an AI can unleash your underwriters’ true potential? Liberate provides everything you need to build end-to-end automations that give your underwriters and claims adjusters superpowers. Learn more.