Auto logistics transporters and car rental companies may serve different types of fleet customers but have at least one ordeal in common.
If the returned or delivered vehicle has undocumented damages, whose fault, is it? The customer? The service? Or the lack of a complete record due to oversights by one or the other or both?
Artificial intelligence (AI) may finally resolve these routine and vexing disputes that can result in unwanted payments, absorbed costs, or mediation. Each side dreads these options.
In the last few years, AI has boosted tech-driven inspection tools to the point where they surpass the human eye and brain in the speed and accuracy of noting flaws, dings, dents, scratches, and whatever other anomalies seem to mysteriously emerge.
AI Tool Cuts Inspection Times 60%
Ship.Cars, a nationwide platform that connects vehicle shippers, carriers, and owner-operators, has used an AI inspection damage tool since June 2022 saving time and hassles like never before.
The logistics provider handles about 20,000 to 30,000 inspections per month throughout the U.S. at pick-up and delivery locations to spot any damage during transport.
The AI tool reduces vehicle inspection times by up to 60%, with full condition reports sent within 30 seconds, said Jonathan Stott, senior vice president at Ship.Cars.
Ship.Cars partnered with Monk AI which specialized in the damage inspections space, he said.
“Before all this tech, everyone was doing it with paper, pen, and walking around the car, asking the customer to verify the information, and then driving off,” Stott said. “That has a lot of potential for user error and liability.”
One more advanced inspection method still used is the driver marking flaws on a digital screen with pictures, which helps lower the liability but doesn’t spot all the problems.
AI + Video = More Detail
Building on the technology of taking photos and video, AI algorithms use the media to find damages automatically, with AI distinguishing among dents, dings, and scratches.
“Now with all-around video, the tool can look at the video and mark its flaws, and then drivers only add or remove damage 5% of the time,” Stott said. “We still do the final review and the customer reviews [the vehicle] as well.”
The AI can draw on smartphone video camera systems, both Android and Apple.
Stott outlined a typical process: “The driver gets to the location on a side street or parking lot, contacts the customer who hands off car and keys, and will walk around to do the inspection, he said. “The customer will review and then sign off. Then he rolls the vehicle onto the trailer.”
The reverse process happens at the destination.
With the enhanced digital inspections, the AI can map the information onto the existing format of the report. “It looks the same as before but adds the marks,” Stott said.
The AI tech belongs to Monk AI, and through the partnership and integration, transforms the inspection report. “It decodes the video and then transfers it to our technology,” Stott said. The system finds an average of three damages per inspection.
Compared to paper inspections, the AI system saves a lot of time and lowers liability risk considerably, he said. Most of the benefit comes from a decrease in risk which is valuable but hard to quantify.
Finding More Vehicle Damages
In the auto rental industry, certain companies are deploying an involved application from Ravin AI that can help rental car operations recover on more damages and improve the customer experience, said Eliron Ekstein, co-founder and CEO of Ravin AI. “A lot of damage goes undocumented with car rental companies.”
About 40% of rental fleets are carrying some sort of damage, Ekstein said. Much of it is not documented when renting a vehicle because customers don’t take enough time to inspect the vehicle.
“If damage is not documented, you could be the one paying for it when you return the vehicle,” he said. “Car rental companies don’t want to charge you for something you haven’t done, but they have no choice when they haven’t found the villain in good time.”
Ravin AI offers a system that uses cameras positioned on either roads or mobile devices that can capture vehicle conditions at that point in time. It creates a report of damage and can identify if it was incremental. It draws on six years of vehicle analysis through AI and computer vision.
Ravin AI technology deploys its own patented and proprietary technology. It’s used by clients in the U.S., Europe, Japan, and Australia. OpenLane has invested in Ravin AI as a strategic partner.
At Heathrow International Airport in London, for example, four cameras at the entrance to the Avis/Budget rental car facility demonstrated over 22% more damage than previously documented, by seamlessly scanning vehicles as they drive through. The algorithm decides with angles to use, Ekstein said. “This is critical to rental because you don’t want to interrupt the flow.”
In addition to Avis and Hertz, Ravin AI is working with Buggy TLC Rentals in New York to phase in the technology.
Creating a Complete Condition Report
Ravin AI’s remarketing clients, such as Toyota, will do scans for wear and tear and create a full condition report for remarketing channels. A complete CR includes images of the vehicle from multiple exterior angles, the interior, and measurements of tire/tread depth.
Ravin AI can extend its platform from damage assessments to a full condition report.
Conventional damage and auto grading systems still rely on manual human input that can lead to varied assessments and mistakes, Ekstein said.
“If you sell car based on a 3.6 [score] because that’s what the inspection company gave you, and someone else sells it at 4.2, you feel like you’ve left a lot of money on the table,” Ekstein said. Every point can add or subtract thousands of dollars from the resale value of a vehicle, he added.
Ravin AI standardizes condition reports as much as possible and creates enough data to make inspection reports less disputable and reduce arbitrations.
Another advantage is speed. The app inspects the vehicle, gets the market price, scans for potential buyers, and posts the vehicle on the most suitable online marketplaces — all in the same day. Its overall efficiency cuts the process from weeks to minutes, Ekstein said.
He outlined the old system sans AI: If a customer returns a leased vehicle to a dealer, the dealer will then order an inspection company, the inspector travels to the dealership, and then writes up a condition report on a notebook or tablet. Then the inspector sends an estimate and grade to the dealer and leasing company. The dealer would decide whether to price it, and if so, where and when to place the vehicle for sale. That process can take days and weeks.
“By using the app on the spot, they don’t need to coordinate and wait for anyone, and the customer can get the vehicle [transacted] faster,” Ekstein said.
One Company Upping Its AI Game
Digital imaging provider Black Widow has emerged from its start-up phase and is creating its latest version of digital AI vehicle damage inspections.
In the past year, new owners acquired Black Widow and opened an R&D facility in Indianapolis in November 2022 as they moved beyond a Chapter 11 bankruptcy.
Black Widow’s AI components use image enhancements, cropping, and background replacements as it highlights vehicle features automatically while growing and learning with AI, said Darren Kemper, president of Black Widow Imaging.
“We really want to be an agnostic platform and work with various damage detection companies out there in the market,” Kemper said. “There’s a lot of change coming in few years.”
Black Widow has integrations with PAVE, an advanced virtual vehicle inspection platform, and Click-Ins, which provides AI-powered vehicle inspections. Black Widow is in talks with other companies about the early stages of integration and getting to market.
Black Widow clients include 20 auctions, six dealer locations, two boating OEMs, and an RV OEM.
In the auto auction industry, the end game is to leverage AI to supplement the CR process and get the time down, Kemper said. With CR time averaging 15-25 minutes per vehicle, any way to speed that up and build trust and confidence in CR is desperately needed and sought after, he added.
Refining the Digital Image Technology
Black Widow, based in Chesterfield, Missouri, plans to continue refining the hardware needed for photos and streamline the process. “Our focus this past year has been revamping and modernizing client equipment and the server side for image processing and syndication.”
With three active patents and three pending ones, the Black Widow R&D facility has explored how to modernize the platform at customer locations. “We’ve taken it from a PC-based software platform and moved it to a mobile app platform to slim down the footprint at the client location. It’s all controlled through a mobile app on a tablet or phone.”
Black Widow plans to revamp its “black spider” camera structure and release a new version in coming months. It will include an enhanced lighting system and more cameras.
“A big challenge for the auction industry is getting information mapped over to existing systems at auctions,” Kemper said. “In the next 12 months, there’ll be a lot of movement on getting data feeds piped through auction management system platforms. That is the next big step in getting the plumbing put together.”
Outside of auctions, AI offers opportunity for the rental car industry, given their large lots and recurrent damage disputes from customers, Kemper said. It eventually plans to position its product for auto rental since the companies need to inspect vehicles continuously leaving and returning.
What’s Next for AI Damage Finders
Ekstein foresees more advances with AI damage inspection technology as it rapidly changes and improves its use of data produced by mobile and CCTV cameras from real environments.
“AI will become more sophisticated in filtering out dirt and terrible lighting and imperfect filming by ordinary people holding a phone,” he said. “It can intervene when doing an inspection to capture the right images and give feedback and expand into other parts of the inspection.”
AI will even evolve to the point where it can suggest repair estimates and recommend whether to replace a vehicle, he said, like a virtual quote from a body shop. That only boosts the credibility of damage and condition reports.
“The potential use cases are tremendous,” he said. “When exploring a trade appraisal, you always think a dealership is lowballing you on the price of the car. With this, you can push back against the ‘mechanic in the back pocket’ of the dealer,” and use it to make a [fairer] trade.”
As applied to vehicle insurance policies, the Ravin AI inspection technology can undergird more accurate claims and repair assessments for damaged vehicles.
“You can give the policy holder and the customer the estimated costs for repairs and help the customer decide what to claim and help the insurance company decide which body shop to use.”