How computer vision reduces claims cycle time

Streamlining the claims process is a major goal of the insurtech movement, and artificial intelligence is making it a reality.

Ageas is among the insurers working with the startup Tractable's computer-vision platform that can assess damage to autos and homes. Founded in 2014, the startup, with offices in San Francisco and London, is refining its algorithm for a wide range of clients including insurance companies and collision-repair providers. It has secured nearly $10 million in total equity funding, including $8 million from a Series A fundraising round in June 2017.

Adrien Cohen, co-founder and chief commercial officer of the company says that using AI allows insurers to review more claims faster, reducing cycle time for customers.

Cars sit piled up on the side of the road in the Rockaway area of New York.
Cars sit piled up on the side of the road in the Rockaway area of New York, U.S., on Wednesday, Oct. 31, 2012. Atlantic superstorm Sandy may cut U.S. economic growth as it keeps millions of employees away from work and shuts businesses from restaurants to refineries in one of the nation’s most populated and productive regions.

“Most insurance companies only review 30% to 40% of the claims that come in,” said Cohen. “With AI, we can review 100% of the claims.”

Computer vision isn't just limited to just car insurance claims, Cohen adds, but can be used with any type of P&C loss.

“Any type of damage is assessed visually, and that’s something AI can learn,” he said. “The entire process can be streamlined if it is dealt with by computers instead of humans.”

Cohen said one of the biggest technological challenges the company faces is supervising the learning of the algorithm, and finding experts who review the data and are able to transfer human knowledge into machine learning.

For reprint and licensing requests for this article, click here.
Artificial intelligence Claims Insurtech Machine learning
MORE FROM DIGITAL INSURANCE