Ericson Chan, Zurich Insurance’s Chief Information and Digital Officer, recently announced that Zurich is starting to experiment with OpenAI’s natural language processing tool ChatGPT in claims and underwriting.
AI can create “a huge amount of efficiency” in jobs such as extracting information from long documents and writing code for statistical models, he told the Financial Times. Zurich UK already uses our AI technology to settle claims in real time.
The arrival of ChatGPT has prompted many in the insurance industry, and further afield, to wonder – and worry – whether AI will replace humans.
Here at Sprout, we have always been convinced that AI is not here to take humans’ jobs, but help them offload repetitive tasks. Our AI-powered claims processing software takes jobs like reading documents and inputting data off claims handlers’ plates, so they can focus on high level work and an empathetic, differentiated customer service.
“The idea that AI will replace people altogether, has been a broad misunderstanding of AI in the Insurance Industry up to now. The hype around ChatGPT has raised awareness of the potential of AI. During claims processing, AI can efficiently and accurately sift through and review data so claims handlers have time to provide a better customer experience instead of processing data.”Sprout’s CEO Roi Amir
Chan agreed. “For underwriting, for claims, it is not going to replace [people] but it is going to make it a lot more efficient,” he said.
This blog will explore the ways our AI helps claims handlers work more efficiently and focus on customers and resolving complex claims.
Automate repetitive tasks
When claims processing is automated, claim handlers can devote more time and attention to customers since they are freed from the time-consuming tasks of collecting and reviewing claims information, overseeing payments, and updating customer information. As a result, they are able to respond to customers’ inquiries more promptly and empathetically, with the ability to assure them that their claims will resolved efficiently.
Lower the risk of bias or error
AI-powered claims processing minimises the risk of human error or bias, ensuring a fair and just outcome for insurers and customers alike. This approach enables insurers to process large volumes of claims data at scale and high speeds, providing customers with the assurance that their claims are being handled accurately, and handlers with the assurance that outcomes are fair.
Process claims faster
People nowadays expect streamlined and straightforward services, whether they are shopping, banking – or making an insurance claim. That expectation may not be met without automated claims processing. Our AI helps insurers process claims in real time, resulting in satisfied customers and a higher Total Net Promoter Score (TNPS).
How Sprout.ai helps claims handlers work more effectively
Our patented technology can process claims forms and extract the meaning of a claim far faster than a human. This is particularly useful in a world where the volume of data is increasing exponentially.
Optical character recognition (OCR) capabilities can read handwritten language at a level that surpasses human competence, in languages from Greek to Japanese. It can help eliminate loss making or incorrect decisions by automating the process and referring to thousands of data points.
We partner with insurers, integrating data-led technology with their claim management systems and processes for a tailored, individualised approach. Our open API enables automation and digital transformation through insurers’ existing systems to ensure our AI meets insurers’ distinct needs.
“The pilot was great fun and I learned some things along the way. The Sprout.ai team was second to none. They were fully invested and listened to our feedback to ensure it’s a success for all. I will certainly be singing their praises.”Zurich claims handler after working with Sprout.ai
How automated policy checking AI works
- Firstly, historical data, such as claim types, policies, and settlement decisions are analysed.
- A library of hypotheses for coverage inclusions and exclusions based on that historical data is then built. For example, “forced entry to the car” would be included, but “car’s doors were unlocked” would be excluded depending on the policy wording.
- At Sprout.ai, our AI then processes the claim description and context, retrieves the policy information, and identifies the relevant policy sections. Natural language processing (NLP) and Optical Character Recognition (OCR) can contextualise policy language to turn that language into actionable decision making.
- Policy checking then compares the claim description to policy coverage and provides a recommendation on coverage and settlement amount.
- The conclusions are returned to the claims platform and claims handler.
Sprout.ai was developed using more than 20,000 historical claims. Every new claim added to the platform brings new knowledge and experience. As with a human claims handler, this will help the AI to keep getting better at what it does. The process reduces errors, increases fairness and frees handlers from repetitive tasks so they can focus on customer service and complicated claims.