The system is make and model-agnostic

The Hartford, one of the oldest property and casualty insurance companies in the US, plans to use AI to speed up car insurance claims by automatically analyzing photos of vehicle damage.

The AI-based solution was created by five-year-old British startup Tractable, which develops computer vision systems for the insurance sector that can estimate repair costs in real time, and claims similar results to expert human appraisers.

Immediate results

“We are honored to announce this partnership,” Alex Dalyac, co-founder and CEO of Tractable, said. “We see this as a testament to the quality of our technology, the maturity of our solution, and the trust in our people. Computer vision is accelerating accident recovery; the technology is here, it’s on the ground and it’s making a difference.”

Tractable’s AI has been trained on “many millions” of images to assess damage by analyzing multiple photos. Aside from vehicle damage assessments, the system also can analyze destruction caused by natural disasters, by taking feeds from satellite, drone, or smartphone imagery to provide a holistic view of assets.

“Using AI in this way can increase the speed of remotely reviewing claims from days to minutes, removing inefficiencies from the process, helping insurers and repairers agree on repairs faster, and accelerating accident recovery for drivers,” states the announcement.

Connecticut-based Hartford was established in 1810.

Tractable, founded in 2015, has raised $55 million in venture funding, and its technology is used to process auto claims for insurers including Tokio Marine in Japan, Covea in France, the European insurer Admiral, and for US auto-parts provider LKQ.