Atrium Underwriters is to automate the initial evaluation stage of its North American property claims through a partnership with IT firm Tautona.
The partnership will use an AI system built by the latter, and which the companies say has been through a ‘lengthy pilot stage’. It is understood that it is already being used for live claims.
Jonathon Hopper, claims adjuster at Atrium Underwriters, said the company was focused and delivering faster, more-consistent claims for brokers and clients through the partnership.
He added: “In doing so we are delivering on the Lloyd’s Blueprint 2’s priority to reduce the claims lifecycle and it is extremely positive to be working in partnership with Tautona, which is a graduate of the Lloyd’s Lab. Atrium prides itself on the quality of its claims management and strongly believes that this significant investment will achieve improved outcomes across our North American property claims.”
Tautona appears to use a system known as Natural Language Processing that scans and searches documents for keywords and phrases that it uses machine learning to draw conclusions from.
The official Lloyd’s website says of Tautona: “The data pool for London market claims is very deep, and there are sometimes dozens of linked policy documents that must be analysed – previously always by a human claims handler – to understand the technicalities of a single claim. As a result, the automation of P&C claims checking was once thought to be impossible, and few firms were willing to even attempt it.”
It goes on: “Before entering the Lab, Tautona had already developed an early-stage claims adjustment tool, designed to read each claim and its linked policy documents to produce a recommendation on whether a claim should be paid. The product was conceptually right, and already in use in other industries, as well as in consumer insurance. The challenge was to bring it into the P&C insurance market successfully.”
As part of its development, Tautona was matched with Atrium on a ten-week pilot programme through which the two companies were able to create a proof-of-concept model showing how AI could reduce costs and processing times without losing accuracy.
Says Lloyds: “The claims tool that was subsequently rolled out has been trained on hundreds of thousands of claims and reads them as a human would, but more accurately and within seconds, instead of 2-3 hours. Tautona says that while human beings analyse claims at 95-96% accuracy, their system will deliver 98% accuracy.”
