Much ink has been spilled on the promises and failures of AI's highest-profile automotive application: self-driving cars. But the much-hyped technology could reshape another aspect of the automotive experience: buying a car. Executives at Cox Automotive, speaking at a Detroit event last week, expressed optimism about the ways they believe AI could make a notoriously time-consuming and difficult process easier on both buyers and sellers. Take a look: Already, Manheim, the Cox-owned wholesale vehicle auction, is deploying AI to collect images and identify damage on vehicles. Manheim has 23 fixed imaging tunnels installed at 13 auction sites, with plans to double the equipment by year's end, Brad Burns, Manheim's AVP of vehicle information, told Tech Brew. Vehicles drive through the tunnel and, in seconds, the equipment captures 2,500 images, including of the vehicle's undercarriage. AI selects the 12 best images. The rest are run through a machine-learning model that can detect damage, speeding up the process of conducting an inspection and generating a vehicle condition report, Burns explained. "Right now, for the most part, the entire process is manual," Burns said. "As we're starting to deploy this technology, certain aspects of it will be presented to [inspectors] just to validate…Subjectivity is a big challenge when it comes to describing the condition of the vehicle because everyone has a different set of eyes." Keep reading here.—JG |
No hay comentarios:
Publicar un comentario