How Insurers Are Applying AI

Insurers should not invest in technology-driven projects; instead, look for use-case-driven projects.

AI is everywhere. Insurers are piloting various AI projects, insurance technology vendors are building it into their solutions, some insurtech startups are all AI-powered and horizontal tech vendors are creating AI platforms that sit underneath it all. Insurers that haven’t experimented with AI yet are benefiting from the technology through third-party relationships, even if they don’t realize it. 

Unfortunately, the broad scope covered by the umbrella term “AI” can cause confusion for insurers — especially because some technology providers use this label to better position their offerings in the marketplace.  

Usage of AI in the insurance world can typically be broken down into four categories:  

  • Machine Learning. The goal of machine learning, a process where an autonomous system learns from a data set to identify novel patterns, is often to refine underwriting or claims algorithms. Applications include advanced predictive modelling and analytics with unstructured data. 
  • Image Recognition. Until recently, images were a type of unstructured data better resolved by humans. Image recognition leverages AI to extract insights from digital image analyses. Applications include photo analysis and handwriting processing. 
  • Audio Recognition. AI-enhanced audio recognition captures any sound (from human speech to a car horn) and turns it into a rich, usable data source. Applications include speech recognition and non-voice audio recognition. 
  • Text Analysis. AI-powered text analysis is pulling out meaningful insights from a body of text (structured or unstructured). Applications include form reading and semantic querying. 

Justifying the Use of AI in Insurance

Novarica’s Three Levers of Value framework can help conceptualize the business value of each AI use case for insurers. Each of these levers — Sell More, Manage Risk Better and Cost Less to Operate — is applicable to a specific AI technology use case. 

Helping insurers identify upsell/cross-sell opportunities, for example, falls under sell more, while accelerating underwriting risk assessment could be categorized as managing risk better and enabling more efficient help desk support helps insurers cost less to operate. 

These are just a few examples of the value AI can bring insurers; AI use cases span categories such as product/actuarial, marketing, underwriting, customer service, billing, claims and compliance. Key use cases include: 

  • Deploying better pricing models. This machine learning use case chiefly falls in the domain of product owners and actuaries, as it applies to the area of predictive analytics. In this case, AI can help actuaries make better decisions when pricing products, thus managing risk better. 
  • Improving marketing effectiveness. This machine learning marketing use case involves using third-party or internal tools to analyze vast amounts of raw data and identify the media channels and marketing campaigns with the greatest reach and engagement levels. Here, big data analytics can help insurers sell more. 
  • Performing better property risk analysis. Using AI-powered photo analysis, underwriters can generate faster and more accurate roof damage estimates. Ultimately, this helps insurers manage risk better. 
  • Leveraging smart home assistants to deflect calls from call centers. Through a voice prompt to their smart home assistants, customers can get quotes, request policy changes and even start a home insurance claim thanks to AI-powered audio recognition. By offering another avenue to help answer customers’ FAQs, insurers free their call center employees to address more complex customer inquiries, decreasing operating costs. 
  • Increasing invoice processing speeds. Through use of text analysis and image recognition technology, AI can help billing staff eliminate error-prone human invoice handling. Using AI-powered form reading leads to greater process efficiencies, which lowers operating costs. 
  • Identifying and mitigating claims fraud. Here, machine learning can help identify potentially fraudulent claims faster. This processing speedup gives claims staff more time to focus on higher-value transactions and leads to better risk management. 
  • Enabling automatic handling of compliance requirements. Machine learning can help team members improve compliance and reporting by automatically handling complex compliance requirements. This results in lower operating costs as compliance staff can direct their attention to tasks requiring human review. 

See also: 4 Post-COVID-19 Trends for Insurers

The AI ecosystem is evolving quickly, with new technology applications emerging every day. We may soon even see further AI and ML processing speedups with the advent of quantum artificial intelligence and machine learning.  

Insurers should not invest in technology-driven projects; instead, governance should search for use-case-driven projects that most benefit the company. However, in the case of important emerging technologies — like AI and ML — it’s valuable to look for ways to deploy that technology and build up skill sets (and culture) within the organization. Additionally, many insurers have an innovation group whose sole purpose is to future-proof the organization by seeking out opportunities to deploy emerging technologies. In these cases, it’s important to refer to actual business use cases and elucidate the concrete value they provide to specific business units.

To learn more on this topic, check out Novarica’s brief, Artificial Intelligence Use Cases in Insurance.


Jeff Goldberg

Profile picture for user JeffGoldberg

Jeff Goldberg

Jeff Goldberg is head of insurance insights and advisory at Aite-Novarica Group.

His expertise includes data analytics and big data, digital strategy, policy administration, reinsurance management, insurtech and innovation, SaaS and cloud computing, data governance and software engineering best practices such as agile and continuous delivery.

Prior to Aite-Novarica, Goldberg served as a senior analyst within Celent’s insurance practice, was the vice president of internet technology for Marsh Inc., was director of beb technology for Harleysville Insurance, worked for many years as a software consultant with many leading property and casualty, life and health insurers in a variety of technology areas and worked at Microsoft, contributing to research on XML standards and defining the .Net framework. Most recently, Goldberg founded and sold a SaaS data analysis company in the health and wellness space.

Goldberg has a BSE in computer science from Princeton University and an MFA from the New School in New York.

MORE FROM THIS AUTHOR

Read More