With a system from Tractable AI that has processed over $1 billion in auto claims to date

Covéa, France’s largest insurance group, is introducing an AI-based system to analyze damage to claimant’s cars.

The system, developed by London-based startup Tractable, uses computer vision to understand collections of photos of damaged vehicles, making sense of the impact as a human would.

It will be deployed in bodyshops across France, where it will assess claims for three insurance brands – MAAF, MMA, and GMF.

Accelerating claims with AI

Tractable said its computer vision systems have processed over $1 billion in auto claims for over 20 of the world’s top insurers, including Ageas in the UK, Tokio Marine and MS&AD in Japan, and PZU and Talanx-Warta in Poland.

Tractable’s platform relies on deep supervised and semi-supervised learning. The models are trained on “many millions” of images, striving to to come up with a single, accurate assessment from multiple pictures.

"Using AI to accelerate claims assessment is a step-change for policyholders of the MAAF, MMA, and GMA brands, and our repair service providers. In addition to making claims handling more efficient, Tractable’s AI will allow all stakeholders to make better, more accurate decisions that benefit everyone,” Bruno Lacoste Badie, expertise and indemnity solutions director at Covéa, said.

Covéa has worked with Tractable since 2016. During this time, the pair set up the ExcellIA partnership, which includes the service platform called Darva.

Adrien Cohen, co-founder at Tractable, told AI Business, “To develop a leading solution, you need to find a visionary customer, the partner who wants to go on the journey. Not every insurance company wants to be a pioneer and those are not the first customers to have. In our case, we were fortunate to have Covéa championing our technology, which is why we are where we are now in France today.”